BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
Genome-wide association study of perioperative myocardial infarction after coronary artery bypass surgery
ForJournal: peerBMJ Open review only Manuscript ID: bmjopen-2014-006920
Article Type: Research
Date Submitted by the Author: 14-Oct-2014
Complete List of Authors: Kertai, Miklos; Duke University, Department of Anesthesiology/Cardiothoracic Li, Yi-Ju; Duke University, Department of Biostatistics and Bioinformatics; Duke Uniersity, Duke Molecular Physiology Institute Ji, Yunqi; Duke University, Department of Biostatistics and Bioinformatics Alexander, John; Duke University, Department of Medicine/Cardiology; Duke University, Duke Clinical Research Institute Newman, Mark; Duke University, Department of Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute Smith, Peter; Duke University, Department of Surgery/Cardiothoracic Joseph, Diane; Duke University, Duke Clinical Research Institute Mathew, Joseph; Duke University, Department of Anesthesiology/Cardiothoracic Podgoreanu, Mihai; Duke University, Department of http://bmjopen.bmj.com/ Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute
Primary Subject Genetics and genomics Heading:
Secondary Subject Heading: Cardiovascular medicine, Surgery, Epidemiology
Myocardial infarction < CARDIOLOGY, SURGERY, Adult intensive & critical on October 1, 2021 by guest. Protected copyright. Keywords: care < ANAESTHETICS, Anaesthesia in cardiology < ANAESTHETICS, Cardiac Epidemiology < CARDIOLOGY, GENETICS
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from ; Peter K. K. ;Peter 1,6 ; ; ; Mark F. Newman, MD Newman, F. ;Mark 1,6 2,6 Department of Biostatistics and Bioinformatics; Bioinformatics; of Biostatisticsand Department 4 : 2,895
; Mihai V. Podgoreanu, MD Podgoreanu, V. ;Mihai 1 Cardiac Surgery; Surgery; Cardiac 3 ; John Alexander, MD, MPH MD, JohnAlexander, ; BMJ Open 4 http://bmjopen.bmj.com/ Cardiology, and and Cardiology, ; Yunqi Ji, PhD ;Yunqi 2 4,5 Duke Clinical Research Institute; Duke University, Durham, NC, USA USA NC, Durham, University, Duke Institute; Research Clinical Duke on October 1, 2021 by guest. Protected copyright. 6 ; Joseph P. Mathew, MD, MPH MD, JosephP. Mathew, ; 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml ; Yi Ju Li, PhD ;Yi Ju Li, 1 For peer review only cardiopulmonary bypass, genome wide association study, inflammation, myocardial infarction, surgery infarction, inflammation, myocardial study, association bypass, genome wide cardiopulmonary ; Diane Joseph, MA Joseph, ;Diane 3 Divisions of Cardiothoracic Anesthesiology, Anesthesiology, Cardiothoracic of Divisions Physiology Institute; Molecular Duke Duke University Medical Center, Durham, NC 27710. 27710. NC Durham, Center, Medical University Duke Email: [email protected] 919 681 4720; Telephone: 919 681 4776; Fax: Words: Key for the Duke Perioperative Genetics and Safety Outcomes (PEGASUS) Investigative Team Investigative (PEGASUS) Outcomes Safety and Genetics Perioperative Duke the for Smith,MD Genome wide association study of perioperative myocardial infarction after coronary artery bypass surgery bypassartery surgery coronary after infarction myocardial perioperative study of association Genome wide PhD MD, Kertai, MiklosD. 1 5 of Anesthesiology, MD; Department 3094, Box MihaiPodgoreanu, to V. Correspondence references) and tables, figures abstract, title page, (excluding Count Word Page 1 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 2 of 51
6 in stage II. A secondary analysis using CK MB as a a as CK MB analysis using secondary A instage II. 5 2 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. for rs609418), or a dichotomized phenotype based on extreme CK MB values = p values (min 7.72x10 extreme CK MB on based phenotype dichotomized a or rs609418), for 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We conducted a stage I genome wide association study (GWAS) in 1,433 patients (112 cases/1,321 controls) controls) cases/1,321 (112 in1,433 patients (GWAS) study association stage a genome wide I conducted We Incidence of myocardial infarction (MI) after cardiac surgery can be as high as 19%, with considerable implications for implications for 19%, considerable with as high be can as surgery cardiac after infarction (MI) myocardial of Incidence Using 2 stage GWAS and pathway analysis, we identified and prioritized several potential susceptibility loci for loci potential susceptibility several and prioritized we identified analysis, pathway and GWAS 2 stage Using increased mortality and long term morbidity. Identifying populations at risk for developing infarcts, or alternatively, with intrinsic intrinsic with alternatively, infarcts, or developing risk at for populations Identifying morbidity. long term and mortality increased common novel toidentify sought We gaps. knowledge high priority remain agent protective toa responsiveness or cardioprotection, bypass grafting. artery coronary in undergoing MI patients postoperative with associated variants genetic and Results: Methods for rs4834703) supported these findings. Pathway analysis revealed that genes harboring top scoring variants cluster in pathways of pathways in cluster scoringvariants top harboring that revealed genes Pathway analysis findings. supported these rs4834703) for Conclusions: perioperative MI. MI. perioperative STUDY THIS OF LIMITATIONS AND STRENGTHS ABSTRACT ABSTRACT Background:
quantitative trait (min p = 1.26x10 = p (min trait quantitative from the Genetics of Myocardial Adverse Outcomes and Graft Failure (Gene Magic) study, and a stage II analysis of 2,055 patients 2,055 analysispatients of a stage II and study, (Gene Magic) Failure Graft and Adverse Outcomes Myocardial of Genetics the from (225 cases/1,830 controls) combined from the Gene Magic and Duke Perioperative Genetics and Safety Outcomes (PEGASUS) (PEGASUS) Safety Outcomes and Genetics Perioperative Duke Gene Magic and the from combined controls) cases/1,830 (225 control and quality day. Following postoperative first during the ≥10xULN CK MB values as defined was MI Postoperative studies. adjustment for clinical covariates, we identified 521 single nucleotide polymorphisms (SNPs) in the stage I GWAS analysis. Among Among analysis. GWAS I stage polymorphismsinthe (SNPs) nucleotide 521 single we identified covariates, clinical for adjustment 10 < p met regions intergenic or genes in3 variants common 8 these, biologic relevance to extracellular matrix remodeling, ER to Golgi transport and inflammation. transport inflammation. and ER to Golgi matrix remodeling, toextracellular relevance biologic 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
1
2 3 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. However, although family based methods are not practical for studying perioperative MI, its genetic MI, its genetic perioperative for studying practical not methods are family based although However, 3 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only The evidence for heritability of MI is striking, supported both by family studies and more recently, by a number of well of number a by recently, more studies and family by both supported ofis striking, MI heritability for evidence The We conducted comprehensive and complementary single marker and pathway based genome wide association analyses. analyses. association pathway based and genome wide marker single complementary and comprehensive conducted We This is the first genome wide association study of perioperative myocardial infarction, using prospective cohorts of cardiac cardiac of cohorts prospective using myocardial of infarction, study perioperative association genome wide first is the This factors. risk non genetic and adjustment full phenotype, for primary the of definitions standard patients, surgical The study is powered to detect relatively large effect sizes. large effect relatively to is powered detect study The Predominantly Caucasian cohort, thus findings cannot be generalized to other populations. populations. toother cannot generalized be findings thus Caucasian cohort, Predominantly Rare genetic variant effects not analyzed. analyzed. not effects variant genetic Rare Despite advances in surgical techniques and pharmacologic therapy, the incidence of myocardial infarction (MI) after coronary coronary after (MI) infarction of myocardial the incidence therapy, pharmacologic and techniques insurgical advances Despite
• • • • • Identification of populations at risk for developing large infarcts or, alternatively, with predictors of a cardioprotective state, state, or cardioprotective of a predictors with alternatively, or, infarcts large developing for populations risk at of Identification knowledge gaps. high priority remain protective agent toa responsiveness
locus in multiple racial groups. inmultiple racial locus INTRODUCTION artery bypass grafting (CABG) remains as high as 19%, and is associated with increased mortality and long term morbidity. long term mortality and increased with is associated and high 19%, as as remains (CABG) bypass grafting artery basis is strongly suggested by several observations, including wide variability in incidence and severity that is poorly explained by by explained is poorly that severity and inincidence variability wide observations, including several by suggested is strongly basis powered and replicated genome wide association studies (GWAS), which primarily implicate common genetic variants at the 9p21 9p21 the at variants genetic implicate common primarily which (GWAS), studies association replicated genome wide and powered Page 3 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 4 of 51 Given the polygenic polygenic the Given 13 15 Thus, the overall influence of of commoninfluence overall the Thus, 8 12 4 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Yet only a few studies have identified allelic associations with altered susceptibility to myocardial ischemia reperfusion ischemia reperfusion tomyocardial susceptibility altered with allelic associations identified few have a only studies Yet 7 Recently, integrated testing of genes involved in the same biological pathway has emerged as an alternative strategy for strategy alternative as an has emerged pathway biological inthe same involved of testing genes integrated Recently, individual genetic background may also significantly modulate the magnitude of postoperative inflammatory response after cardiac after cardiac response inflammatory postoperative magnitudeof the modulate significantly also backgroundmay genetic individual surgery. genetic variation on the incidence of perioperative MI remains poorly understood. understood. poorly remains MI of perioperative incidence the on variation genetic nature of disease susceptibility, this approach is increasingly being used to identify groups of gene variants with shared cellular shared with variants of toidentify gene groups used being increasingly is this approach susceptibility, disease of nature by this strategy, we adopted this study GWAS. of power In statistical the improving also while disease, enrichedfor are that function after MI perioperative with associated variants common genetic toidentify methodology association genome wide employing first injury in cardiac surgical populations, all based on a candidate gene approach. association candidate gene a on based surgical all populations, incardiac injury clinical and procedural risk factors, different racial susceptibility profiles, and results from preclinical animal models. Indeed, models. Indeed, animal preclinical from and profiles, results different racial susceptibility factors, risk procedural and clinical the such as MI, postoperative of pathophysiology implicated inthe pathways biological in found been variability has genetic extensive that suggests state pro inflammatory of a heritability for evidence Mounting surgery. tocardiac response inflammatory acute complex
evaluating the combined effects of multiple genetic variants with small effect size on a disease phenotype. disease a size effect small on with variants multiplegenetic of effects combined the evaluating CABG, followed by pathway based analyses to uncover biological mechanisms of relevance to perioperative MI. toperioperative mechanisms relevance biological of touncover pathway based analyses by followed CABG, METHODS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from as an elevation in plasma level of creatine kinase kinase creatine of level plasma in elevation an as 19
11 12 5 BMJ Open Of those, 1,433 patients met eligibility requirements after applying quality quality after applying requirements eligibility met 1,433 patients those, Of http://bmjopen.bmj.com/ 18 genome wide association study combined with a pathway analysis approach. analysis approach. pathway a with combined study association genome wide 17 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We performed joint two stage a performed We 16 The stage I cohort (discovery cohort) was comprised of 1,493 ethnically diverse subjects who underwent an isolated CABG CABG isolated an underwent who subjects diverse ethnically 1,493 of comprised was cohort) (discovery cohort stage I The expanded by 622 patients of self reported European ancestry, for a total of 2,055 patients. The additional patients underwent CABG CABG underwent patients additional The patients. of total 2,055 for a ancestry, European self reported patients 622 of by expanded an (PEGASUS), Study Outcomes Safety and Genetics Perioperative the of 1997 2006 part as bypass between cardiopulmonary with Center. Medical University Duke at study longitudinal IRB approved control criteria and excluding patients with missing genotypes or phenotypic information. information. phenotypic or genotypes missing with patients excluding and criteria control and nested, cohortwas discovery the which within dataset joint a in analysis variants the genetic top tested stage we II, In with cardiopulmonary bypass for the first time, and were enrolled between 2002 and 2003 in the Institutional Review Board (IRB) Board Review inthe 2003 Institutional and between 2002 enrolled and were time, first the bypass for cardiopulmonary with of Ex vivo Project of sub study multicenter the ancillary (Gene Magic) Failure and Outcomes Graft Myocardial of Genetics approved, (PREVENT IV). via Transfection Engineering Graft Vein MB isoenzyme (CK MB) that was > 10 times the upper limit of normal, as measured by a core laboratory within 24 hours after after hours 24 within laboratory by core a as measured limitnormal, of the times 10 upper > was that (CK MB) isoenzyme MB The study design and reporting of the results follows the recommendations by “Strengthening the Reporting of Genetic Association Genetic of the Reporting “Strengthening by recommendations the follows the of results reporting design and study The (STREGA). Studies” infarction myocardial perioperative Definitionof MI of definition universal tothe defined MI according was Perioperative populations Patient Page 5 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 6 of 51 arbitrarily arbitrarily ). Methods ). seeSupplemental 6 a priori a was analyses for stage I significance Statistical ). BMJ Open http://bmjopen.bmj.com/ method to control for population stratification ( population stratification for tocontrol method 21
18 on October 1, 2021 by guest. Protected copyright. see Supplemental TableS1 Supplemental see For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We used a modified EIGENSTRAT modified a used We defined replication dataset to obviate type I error. In stage II analyses, the same clinical covariate and principal component principal and clinical covariate the same stage analyses, II In type error. toobviate I dataset replication defined 20 Univariate regression analysis was performed to test differences in demographic, clinical and procedural characteristics characteristics procedural and clinical indemographic, totest differences performed analysis was regression Univariate a priori a defined as a 2 tailed p < 0.001, to balance between the overly conservative Bonferroni correction and type II error, given that we had had that we error, given II andtype correction Bonferroni conservative overly the between balance to < p 0.001, 2 tailed as a defined an adjust for population stratification ( stratification population for adjust on the Illumina Human610 Quad BeadChip at the Duke Genomic Analysis Facility. Sample and genotype quality control of data flow of data flow control quality and genotype Sample Analysis Facility. Genomic the at Duke BeadChip Human610 Quad the on Illumina previously as equilibrium, Hardy Weinberg and SNP missingness, relatedness, cryptic check, call gender rates, of assessment included described. PLINK 1.07, assuming an additive genetic model, and including significant clinical covariates and top 10 principal components to components top principal 10 and covariates significantclinical including model, and genetic 1.07, an additive assuming PLINK analysis Statistical surgery, and that was not attributable to an intervening clinical event or pre operative myocardial infarction (adjudicated by the by (adjudicated infarction myocardial pre operative or event clinical intervening toan not that and was attributable surgery, Events Clinical Committee). PREVENT IV controls and quality Genotyping performed was cohorts both in Genotyping procedures. standard using or blood saliva whole was isolated from DNA Genomic Genome wide association analyses performed in the stage I cohort used multivariable logistic regression models implemented in implemented models regression logistic multivariable used inthe cohort I stage performed analyses association Genome wide between patients with and without postoperative MI, and when statistically significant were used to adjust genetic association tests. tests. association used to genetic were adjust significant statistically when and MI, postoperative without patients and with between 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from ). ). quartiles of the CK MB CK MB the of quartiles CK MB quartile, and “1” for for “1” and quartile, CK MB th st and 4 and st Supplemental TableS2 Supplemental 7 BMJ Open http://bmjopen.bmj.com/ ). Aortic cross clamp time, number of coronary artery grafts, and procedures in procedures addition and grafts, artery coronary of number time, cross clamp Aortic ). on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Supplemental TableS2 Supplemental CK MB quartile from the stage II cohort. Logistic regression with the same set of covariates was performed. performed. was set covariates of same the with regression stage Logistic cohort. the II from quartile CK MB th Perioperative MI was observed in 112 of 1,433 patients (7.8%) and in 225 of 2,055 patients (10.9%) in stage I and stage stage and II stage inI (10.9%) patients in225 2,055 of and patients 1,433 of (7.8%) in112 observed MI was Perioperative to CABG were significantly associated with perioperative MI in stage I analyses. Stage II analyses showed that extracardiac extracardiac that Stage showed analyses II analyses. instage MI I perioperative with associated significantly were toCABG ( potential confounders for tests association genetic the toadjust models regression logistic subjects within the 4 the within subjects cohorts, respectively ( respectively cohorts, arteriopathy and year of surgery also played significant roles. All clinical variables identified above were included in multivariable inmultivariable included above were identified variables clinical All roles. significant played also surgery of and arteriopathy year distribution. For CK MB, a linear regression model was applied with adjustment for the same set of covariates as in the primary primary inthe as of set covariates same the adjustment with for applied model was regression linear a CK MB, For distribution. 1 the within subjects for “0” trait as status we re assigned subset, CK MB the extreme For analysis. adjustments (re estimated for the expanded dataset) were applied, with statistical significance defined as a Bonferroni adjusted p value adjusted p value Bonferroni a defined as significance statistical with applied, dataset) were expanded the for (re estimated adjustments RESULTS Finally, we employed pathway analysis to prioritize association results and provide biological interpretation, using functional ontology ontology functional using interpretation, biological and provide results association prioritize to pathway analysis employed we Finally, ). Methods see Supplemental St. Joseph, MI; (GeneGO, toolsinMetaCore analysis implemented enrichment of 0.05/number of SNPs identified in stage I, to control the overall type I error rate. Haplotype association analysis was performed in was performed association analysis Haplotype rate. type error I overall tocontrolthe instage I, SNPs identified of 0.05/number of the stage II cohort for genes tagged by the significant SNPs, adjusting for the same clinical covariates and principal components. components. principal and clinical covariates the same for SNPs, adjusting significant the by tagged for genes cohort stage II the CK MB of the phenotype continuous using cohort, stage II SNPs top in the the identified for were performed analyses Secondary 1 subjects the within whichincluded phenotype, CK MB dichotomized a extreme and values Page 7 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
6 6 6 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 8 of 51 ), 521 521 ), ). ). a member of of member a Figure 1 Figure Table1 SEC24D, SEC24D, 0.46(0.33 0.63) 0.46(0.33 0.63) 2.47 x10 2.43 x10 0.47(0.34 0.65) 4.70x10 4 4 4 ), 534,390 markers were were markers 534,390 ), ), were in high linkage inhigh were ), for 521 SNPs). The top 2 SNPs SNPs 2 top The SNPs). 521 for 5 (OR 0.46; 95% CI, 0.33 0.63), 0.33 0.63), CI, 95% 0.46; (OR 6 PAPPA2 , rs2044061 on chromosome 8), and were and were 8), chromosome on rs2044061 ,
6 ). ). Stage I: Discovery Dataset* Stage ReplicationII: dataset* OR (95% CI) P value OR (95% CI) P value ), 2 in SEC24 family, member D ( D member inSEC24family, 2 ), HDAC4 8 BMJ Open ). A quantile quantile (Q Q) plot of observed vs expected p values expected vs observed plotof (Q Q) quantile quantile A ). http://bmjopen.bmj.com/ Supplemental Figure S2 Figure Supplemental ene ene G location MAF (OR 0.46; 95% CI, 0.33 0.63) and 2.47x10 and 0.33 0.63) CI, 95% 0.46; (OR Intron 0.18 0.39(0.24 0.64) 1.85 x10 Intron 0.17 0.41(0.25 0.66) 2.67x10 Intron 0.18 0.39(0.24 0.64) 1.81 x10 6 Supplemental Methods: A.1. Quality control of flow data Quality control A.1. SupplementalMethods: on October 1, 2021 by guest. Protected copyright. Supplemental TableS10 Supplemental For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml PAPPA2 PAPPA2 For peer review only PAPPA2 defined discovery threshold of p< 0.001 (minimum p = 2.76x10 = p (minimum 0.001 p< threshold of discovery defined a priori a In stage II analyses, 8 of the 521 SNPs met the Bonferroni correction threshold (p < 9.6 x 9.6 < 10 threshold (p correction of 8 Bonferroni the SNPs met 521 stage the analyses, II In After applying quality control criteria ( control criteria quality applying After disequilibrium, with p values for stage II of 2.43x10 of stage for II p values with disequilibrium,
(rs10454444 and rs10913237), located in the pregnancy associated plasma protein A2 gene ( A2 gene plasma protein pregnancy associated inthe located rs10913237), and (rs10454444 SNPs met the the SNPsmet subsequently analyzed in stage ( instage II analyzed subsequently ( for adjusted was well substructure population that showed respectively. In addition, one intronic SNP in histone deacetylase 4 ( deacetylase 4 SNP inhistone intronic one addition, In respectively. analyzed for association with perioperative MI in stage I. While none of the SNPs reached genome wide significance ( significance the genome wide of SNPs reached While none MI I. instage perioperative with association for analyzed
the cytoplasmic Coat Protein II complex), and 2 located in intergenic regions were associated with perioperative MI MI ( perioperative with associated regionswere inintergenic located 2 and complex), Protein Coat II cytoplasmic the Top 8 Single Nucleotide Polymorphisms Associated With Postoperative Myocardial Myocardial Infarction With Postoperative Associated Polymorphisms Nucleotide Single 8 Top 1 rs10913237 176630416 1 rs10489478 176566350 Chr SNP Base pair symbol Gene 1 rs10454444 176649181 Table1. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
6 6 6 6 6 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from , ,
SEC24D SEC24D 1.98(1.47 2.66) 1.65(1.32 2.05) 6.80 x10 0.62(0.50 0.76) 9.55 x10 1.67(1.34 2.09) 8.89 x10 5.81 x10 2.23(1.57 3.17) 8.18 x10 5 4 4 4 4 ) showed the strongest association, association, strongest the showed ) 5 ). Overall, the rs4834703 SNP rs4834703 in the Overall, ). , regulatory factor X associated protein; X associated factor regulatory , (p = 8.26x10 = (p , thyrotropin releasing hormone degrading ectoenzyme hormone degrading thyrotropin releasing , Table2 RFXAP 0.07 2.22(1.42 3.46) 4.34 x10 HDAC4 TRHDE 9 BMJ Open http://bmjopen.bmj.com/ ). When the dichotomized extreme CK MB phenotype was studied inthe was phenotype CK MB extreme dichotomized the When ). Intron 0.10 2.27(1.54 3.34) 3.13 x10 Intron Intron 0.30 1.73(1.29 2.33) 2.59 x10 (rs4834703 and rs6822035, p=0.002) and rs609418 in the intergenic region intergenic rs609418in the and p=0.002) rs6822035, and (rs4834703 Table2 ) and rs10200850 in rs10200850 and ) 6 Intergenic 0.47 0.58(0.43 0.78) 3.51 x10 SEC24D Intergenic 0.26 1.73(1.28 2.33) 3.14 x10 on October 1, 2021 by guest. Protected copyright. (p = 0.0001) 0.0001) ( = (p LOC552889 SMAD9 | | (p = 7.72x10 = (p | | For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml , pregnancy associated plasma protein A2; protein A2; plasma pregnancy associated , DAC4 PAPPA2 SEC24D SEC24D H TRHDE ForRFXAP peer review only , mother against decapentaplegic homolog 9; homolog against decapentaplegic mother , (p=0.001) remained nominally significant ( significant nominally remained (p=0.001) SMAD9 PAPPA2 and and SMAD9 RFXAP When CK MB was tested as a quantitative trait, all 8 SNPs remained significantly associated with plasma levels of CK MB. CK MB. of levels plasma with associated significantly all SNPstrait, 8 remained quantitative a tested as was CK MB When followed by the 2 SNPs 2 in the by followed However, rs4834703 in SEC24D rs4834703 However, stage II cohort, only the in only cohort, SNPs 2 located stage II
consistently showed strong association signals across postoperative MI, quantitative CK MB, and the extreme CK MB traits. CK MB the extreme and CK MB, quantitative MI, postoperative across signals association strong showed consistently between between , histone deacetylase 4; deacetylase 4; histone , 2 rs10200850 240234901 4 rs4834703 119691624 4 12 rs6822035 rs2303970 13 119710421 rs609418 74471340 37417427 surgery, number of diseased vessels, and aortic cross clamp time. time. cross clamp aortic and ofvessels, diseased number surgery, SEC24 related protein D; protein SEC24 related Chr, Chromosome; MAF, minor allele frequency (based on the discovery dataset); OR, odds ratio; 95% CI, 95% confidence interval 95% confidence 95% CI, ratio; odds OR, dataset); discovery the on (based frequency allele minor MAF, Chromosome; Chr, *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of of only, than CABG year other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted HDAC4 Page 9 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
* BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 10 of 51 , , SEC24D 0.78(0.578 1.052) 0.1 1.49(0.950 2.326) 1.82(1.242 2.654) 0.08 1.48(1.152 1.903) 0.002 0.002 1.56(1.189 2.034) 0.001 0.77(0.575 1.043) 0.09 0.77(0.574 1.047) 0.1 CK MB an extreme phenotype as 4 5 6 4 4 4 4 * 0.01 0.84(0.669 1.047) 0.12 P value OR (95% CI) P value 1.21x10 2.27x10 2.26 x10 8.26 x10 7.72 x10 4.81 x10 1.30 x10
4.80 2.93 Β coefficient CK MBa quantitativeas trait , regulatory factor X associated protein; X associated factor regulatory , , thyrotropin releasing hormone degrading ectoenzyme hormone degrading thyrotropin releasing , RFXAP
TRHDE 10 BMJ Open http://bmjopen.bmj.com/ ene ene G location MAF Intron 0.07 9.26 Intron Intron 0.18 Intron 0.10 5.59 0.30 8.54 4.42 Intergenic 0.26 Intron 0.18 5.58 Intron 0.17 5.40 Intergenic 0.47
on October 1, 2021 by guest. Protected copyright. LOC552889 SMAD9 | | | | For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml , pregnancy associated plasma protein A2; protein A2; plasma pregnancy associated , SEC24D SEC24D HDAC4 PAPPA2 PAPPA2 PAPPA2 TRHDE For peer reviewRFXAP only , mother against decapentaplegic homolog 9; homolog against decapentaplegic mother , PAPPA2 SMAD9
, histone deacetylase 4; deacetylase 4; histone , 2 rs10200850 240234901 1 rs10454444 4 176649181 4 rs4834703 12 rs6822035 119691624 rs2303970 119710421 74471340 13 13 rs609418 37417427 1 rs10913237 176630416 1 rs10489478 176566350 Chr SNP Base pair symbol Gene surgery, number of diseased vessels, and aortic cross clamp time using either linear regression (Β coefficient) or logistic regression (Odds Ratio, Ratio, (Odds or regression logistic (Β coefficient) regression linear either using time cross clamp aortic and ofvessels, diseased number surgery, interval) analyses. confidence 95% SEC24 related protein D; protein SEC24 related Chr, Chromosome; MAF, minor allele frequency (based on the replication dataset); OR, odds ratio; 95% CI, 95% confidence interval interval 95% confidence CI, ratio; odds 95% OR, replication the dataset); on (based frequency allele minor MAF, Chromosome; Chr, *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of of only, than CABG year other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted Table 2. inthe Phenotype Extreme an and as Trait Quantitative as a CK MB with Polymorphisms Nucleotide Single 8 Top the of Association Joint Analysis Dataset Dataset JointAnalysis HDAC4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from ). ).
are shown in shown are . Haplotype Haplotype . 10 10
P P × × 5.54 8.84 SEC24D , and , HDAC4 , , Reference OR (95%OR CI) PAPPA2 , OR 1.87; 95% CI, 1.46 2.39, Table3 1.46 2.39, CI, 95% 1.87; OR , 7 0.7126 0.2358 1.87(1.46 2.39) 0.0528 0.86(0.43 1.70) 0.657 0.2874 1.65(1.33 2.06) PMI (n=1830) Patients Patients without
11 BMJ Open Haplotype FrequencyHaplotype http://bmjopen.bmj.com/ gene and results of association tests with incidence of perioperative incidence with tests association of and results gene 0.6044 0.3631 0.3956 (n=225) 0.03516 Patients Patients PMIwith SEC24D on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Table 3 and Supplemental TableS3 shown Supplemental and Table3 in are analysis haplotype of results the and , For peer reviewSNP2 Haplotype only had the most significant association with postoperative MI (p (p MI = 5.54x10 with postoperative association most significant the had Estimated haplotype frequencies in the inthe frequencies haplotype Estimated SNP1 SNPs significant in the regionssurrounding genic structures of haplotype The rs6828577 rs6822035 G C rs6822035 rs10518325 A G rs6822035 rs6822035 rs10518325 rs10518325 C G A A 0.009083 0.02636 1.39(0.67 2.90) 0.378 rs6828577 rs6822035 A A analysis performed for the LD blocks containing the significant markers showed that the A A haplotype (rs6822035, rs10518325) in rs10518325) (rs6822035, haplotype A A the that showed markers the significant containing blocks the for performed LD analysis SEC24D myocardial infarction in the stage II analysis cohort (n = 2055) = (n cohort analysis stage II inthe infarction myocardial Supplemental Figure S3 Figure Supplemental
Table3. Page 11 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 12 of 51
The most significant most significant The
. Reference Reference ). 0.685 0.8592 0.06231 0.76(0.48 1.19) 0.225 0.07845 0.87(0.58 1.31) 0.509
12 Supplemental Figure S4 Figure Supplemental BMJ Open http://bmjopen.bmj.com/ 0.5927 0.9248 0.04425 0.03097 cohorts ( cohorts 8 9 , respectively). The most significant canonical pathway maps in stage I and II comparative comparative and II instage I pathway maps canonical most significant The respectively). , 6 on October 1, 2021 by guest. Protected copyright. and 7.3x10 and For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml and cardiac surgical cardiac and 6 For peer review only 8 rs6822035 rs10518325 C A rs10518325 rs11098451 G G rs10518325 rs11098451 A A rs10518325 rs11098451 A G S5 Figure shown SNPs inare Supplemental the analysis stage I for maps pathway enriched 10 top The OR, odds ratio; 95% CI, 95% confidence interval; SNP, single nucleotide polymorphism; PMI, perioperative myocardial PMI, myocardial perioperative polymorphism; nucleotide SNP,interval; single confidence CI, 95% ratio; 95% odds OR, infarction. than other infarction, procedure myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted time. cross-clamp aortic diseased vessels, of of surgery, number year only, CABG and infarction myocardial associated with SNP previously rs10116277 the including MI, incident postoperative and locus 9p21 the at variation genetic between for association no evidence found we analyses, and II stage I factor adjusted risk multivariate In non surgical both in mortality
were “Cell adhesion: extracellular matrix remodeling” and “Cytoskeleton remodeling: TGF, WNT, and cytoskeletal remodeling” remodeling” WNT, andcytoskeletal TGF, remodeling: “Cytoskeleton and matrix remodeling” extracellular adhesion: “Cell were 1.9x10 = p value (enrichment 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from SNPs in ) ) in single 6 ) ) and 2 intergenic SEC24D (p = 2.43x10 , SEC24D PAPPA2 HDAC4 , , ) ) or as an extreme CK MB phenotype 6 PAPPA2 ) ) in haplotype analyses, and by 7
13 BMJ Open http://bmjopen.bmj.com/ is an integral component of cytoplasmic coat protein complex II (COPII) showed the most consistent strong association with all three phenotypes SEC24D SEC24D on October 1, 2021 by guest. Protected copyright. Supplemental Tables S4 and S5 ). TablesS4 and Supplemental A A (rs6822035, rs10518325, p = 5.54x10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only SEC24D , respectively; respectively; , 3 and 1.1x10 and 4 Among Among these 8 SNPs, rs4834703 in We We present the first report of a 2 stage GWAS involving 225 perioperative MI cases and 1,830 controls. After accounting for regions. regions. The most significant association with perioperative MI was exhibited by rs10454444 in marker marker analyses, and secondary secondary analyses when CK MB was evaluated as a quantitative trait (rs4834703, p=2.43x10 clinical clinical and procedural covariates, we identified 8 significant SNPs mapped to 3 genes ( (rs4834703 and rs6822035, p=0.002). p=0.002). rs6822035, and (rs4834703 analyses were “Immune response: NFAT signaling and leukocyte interaction” and “Cell adhesion: extracellular matrix remodeling” (p matrix remodeling” extracellular adhesion: “Cell interaction” and leukocyte and signaling NFAT response: “Immune were analyses 7.2x10 = DISCUSSION evaluated evaluated in this study. The transport protein transport machinery, a key player in vesicle trafficking of secretory proteins from the apparatus endoplasmic for delivery to downstream reticulum compartments. COPII is responsible (ER) for cargo sorting and to vesicle morphogenesis, with roles the in Golgi modulating ER exit, cell surface transport, lipid secretion and cholesterol biosynthesis, and function of G protein coupled receptors. Conditions of ischemia, oxidative injury, or acute phase response result in ER stress through accumulation of misfolded proteins, which leads to activation of the unfolded protein response (UPR) signaling pathway. If the protective mechanisms activated by the Page 13 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 14 of 51 and limiting IGF 2, its 26 Supplemental Supplemental Figure S6 and with SEC24D being the only isoform implicated in 22 23 involve characteristic skeletal malformations, whereas its
14 SEC24D BMJ Open The related protein PAPPA1 cleaves IGFBP 4 and is activated and http://bmjopen.bmj.com/ ; however, the role of PAPPA2 in cardiovascular biology has not been However, there are no reports of human diseases associated with genetic 30 27 28 25
29 on October 1, 2021 by guest. Protected copyright. ), ), a member of the HDAC family, mediates changes in chromatin structure by removing acetyl Animal models of defects in HDAC4 24 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only ). . . SEC24D groups groups from the core histones, resulting in transcriptional repression. HDAC4 is highly expressed in myocardium, where it plays an important role in the regulation of gene expression and is involved in myocardial cell cycle progression, differentiation, and Supplemental TableS5 Supplemental Histone deacetylase 4 ( of of variants in other IGF system component genes, namely IGF 2 and the IGF 1 receptor ( extracellular extracellular matrix secretion. complete complete disruption results in early embryonic lethality. variation in variation plasma Pregnancy associated protein A2 (PAPPA2) is a metalloproteinase that regulates local insulin like growth factor (IGF) released released from vulnerable atherosclerotic plaques. PAPPA1 has been extensively studied as a cardiovascular risk biomarker for both diagnosis and prognosis of acute coronary syndrome cardioprotective cardioprotective effects following ischemia reperfusion. UPR are insufficient, cells die by apoptosis and autophagy. However, altered expression or function of SEC24 proteins could also explain ER trapping of misfolded proteins under conditions of ER stress, inhibits both IGF 1, with myocardial implications for inflammation linked angiogenesis and repair processes, bioavailability bioavailability by specifically cleaving IGF binding protein 5 (IGFBP 5). In experimental models of myocardial ischemia, IGFBP 5 previously previously reported. Consistent with our single marker analysis findings, GWAS pathway analysis identified significant contributions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from by 31 TGF β plays importantTGF β plays and complex roles 33
15 ). ). Although no direct functional roles are currently attributed to this BMJ Open http://bmjopen.bmj.com/ suggesting functional roles for HDAC4 in modulating perioperative myonecrosis, and their Supplemental Supplemental Table S7 32 on October 1, 2021 by guest. Protected copyright. , , 3 ). ). For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Consistent with this single locus gene association result, the bone morphogenic protein (BMP) For peer review34 only ) ) gene, part of the transforming growth factor beta (TGF β) signaling pathway, which is markedly activated in SMAD9 Experimental HDAC inhibition is associated with a profound reduction in ischemia induced myocardial cell death, 31 Supplemental Figure S7 Figure Supplemental ischemia reperfusion ischemia reperfusion injury. intergenic intergenic region, a query of the Regulome and Haploreg databases reports that rs609418 is located within active regulatory elements (GATA2 transcription factor binding site by ENCODE ChIP seq and altering regulatory motifs in Gfi1 and Mef2 by Position Weight Matrix, triggering triggering preconditioning effects and promoting myocardial repair. Genetic variants in HDACs are important determinants of GWAS pathway analysis (p = 5.2x10 One of the intergenic SNPs associated with perioperative MI (rs609418) is located near the mother against decapentaplegic apoptosis. in regulating post infarction inflammatory responses. In animal models, associated signaling with resolution of inflammation, through repression of cytokine and chemokine the gene synthesis, and protection against SMAD myocardial transcription factors is susceptibility susceptibility to cardiovascular diseases, homolog homolog 9 ( infarcted infarcted myocardium. Members of the TGF β superfamily transduce their signal from the membrane to the nucleus via a distinct receptors combination of transmembrane and downstream effectors – the SMAD proteins. pathway, pathway, which also transduces its signals via a SMAD9 dependent cascade, was identified as one of the top scoring pathways in our potential use in predicting individual patient responsiveness to HDAC inhibition. inhibition. toHDAC responsiveness patient individual predicting in use potential Page 15 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 16 of 51
35 Supplemental Supplemental Results B.6. ) show that, on samplebased current
16 Many different reasons may account for this, including inadequate 8 9 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Of note, most of the genes identified through pathway enrichment analysis in this study encode targets for therapeutic 14 Several limitations are worth mentioning. Power calculations ( limitations Several calculations Power worth mentioning. are locus locus and risk for perioperative MI or mortality after CABG. Strengths of our study are 1) a relatively large population of cardiac surgery patients, 2) a prospective cohort design, and 3) a approach, replication of association findings at the gene and pathway level is much easier compared to replication at the individual SNP level. drugs drugs that have already been developed. Thus, by improving risk assessment and identifying allele specific therapeutic responses, further investigation of loci and pathways prioritized in this cardioprotection. study could yield actionable results for enhancing perioperative size and incidence of perioperative MI, our study can detect a genotypic relative risk ~2 with 80% power (assuming a variant with sample size, variation in study design, differences in allele frequencies, or variability in the definition of perioperative MI phenotype. MI perioperative definitionof inthe or variability frequencies, inallele differences design, instudy size, variation sample Surprisingly, we have been unable to replicate previously reported associations between common genetic variants at the 9p21 combination combination of complementary single marker and pathway based genome wide association analyses. This approach allowed us to identify genetic variants that carry only small disease risk individually, but that jointly can perioperative MI contribute susceptibility. Furthermore, an relatively application of results large from pathway based analysis may effects add structure to on interpreting genomic data and allow exploration of cellular processes that functionally underpin the observed associations. Finally, by using this 10% 10% minor allele frequency and a realistic linkage disequilibrium between the tested marker and the causal locus D’ = 0.8). Thus, although ours is the largest genetic association study of perioperative MI conducted to date, it is powered to detect only common 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from The possibility of rare genetic variants that drive a pronounced
17 8 10 11 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only This work was supported, in part, by National Institutes of Health grants R01 HL075273 and R01 In In conclusion, we present the results of a genome wide association study in a cohort of patients at risk for MI after CABG from targeted cardioprotective interventions. interventions. cardioprotective targeted from FUNDING SOURCES: (to HL092071 Dr. Podgoreanu) and by American Heart Association grants 0256342U and 9951185U (to Dr. Mathew), and 0120492U Podgoreanu). Dr. (to clinical clinical phenotype was not explored in this study, because only variants with minor allele frequencies > 0.05 were assessed. Also, groups. ethnic toother be cannot generalized findings surgery. Based on our integrated approach utilizing both primary (perioperative MI) and secondary phenotypes (quantitative CK MB, extreme CK MB), single marker analysis and pathway analysis, we identified several polymorphisms in the insulin growth factor system implicated in the regulation of extracellular matrix remodeling, as well as the ER to Golgi involved in adaptive responses to ER secretory stress. The functional of significance these novel observations provide may new insights into the pathway potentially variants variants with relatively large effect sizes. Most published genetic association studies of perioperative MI have reported larger effect sizes compared to ambulatory populations (OR range 1.79 3.97). functional functional studies to further elucidate the potential biological effects of the SNPs identified were not feasible due to lack of plasma or tissue availability in the GeneMagic study. Finally, patients enrolled in our cohorts were predominantly Caucasian, and therefore our pathogenesis pathogenesis of myocardial ischemia reperfusion injury, and identification of patients at risk for perioperative MI who would benefit Page 17 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 18 of 51
18 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only None None DISCLOSURES:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (2):483 8, 88 e1 2. e1 2. 88 (2):483 8, 139 (10):1172 9. (10):1172 9. 124
19 (13):1675 84. (13):1675 84. BMJ Open 117 http://bmjopen.bmj.com/ (3):681 9. (3):681 9. (6):585 91. (6):585 91. 305 on October 1, 2021 by guest. Protected copyright. (5830):1491 3. (5830):1491 3. 316 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review (5830):1488 91. only 316 (5):443 53. (5):443 53. 357 artery bypass graft surgery. Jama 2011; surgery. bypass graft artery 9p21.3 and coronary artery disease. Circulation 2008; Circulation disease. artery coronary and 9p21.3 2007; Science 2007; Science Heart, Lung, and Blood Institute Workshop. Circulation 2011; Circulation Workshop. Blood Institute and Lung, Heart, infarction. Science 2007; Science infarction. after coronary artery bypass graft surgery. J Thorac Cardiovasc Surg 2010; Surg J Cardiovasc surgery. Thorac bypass graft artery coronary after 154 2001; Atherosclerosis study. heart family REFERENCES: REFERENCES: coronary following and survival elevation enzyme myocardial of al. et Association V, Hasselblad K, MJ, Mahaffey Domanski 1. 4. McPherson R, Pertsemlidis A, Kavaslar N, et al. A common allele on chromosome 9 associated with coronary heart disease. disease. heart with coronary associated 9 chromosome on allele common A N, al. et Kavaslar A, R, Pertsemlidis McPherson 4. 2. Schwartz Longacre L, Kloner RA, Arai AE, et al. New horizons in cardioprotection: recommendations from the 2010 National National 2010 the from recommendations incardioprotection: horizons al. et New AE, Arai RA, Kloner L, Schwartz Longacre 2. 5. Schunkert H, Gotz A, Braund P, et al. Repeated replication and a prospective meta analysis of the association between chromosome chromosome between of association the prospective meta analysis a and replication P, et Repeated al. Braund Gotz A, H, Schunkert 5. 3. Samani NJ, Erdmann J, Hall AS, et al. Genomewide association analysis of coronary artery disease. N Engl J Engl N Med disease. artery coronary analysis of association Genomewide al. et AS, Hall J, NJ, Samani Erdmann 3. 6. Helgadottir A, Thorleifsson G, Manolescu A, et al. A common variant on chromosome 9p21 affects the risk of myocardial myocardial of risk the affects 9p21 on chromosome variant A, common etA al. Manolescu G, Thorleifsson A, Helgadottir 6. 7. Pankow JS, Folsom AR, Cushman M, et al. Familial and genetic determinants of systemic markers of inflammation: the NHLBI NHLBI the inflammation: markers of systemic of determinants genetic and Familial al. et M, Cushman JS, AR, Pankow 7. Folsom 8. Liu KY, Muehlschlegel JD, Perry TE, et al. Common genetic variants on chromosome 9p21 predict perioperative myocardial injury injury myocardial perioperative predict 9p21 on variants chromosome Common al. et genetic TE, Muehlschlegel JD, Perry KY, 8. Liu Page 19 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 20 of 51 (11 Suppl):I106 12. Suppl):I106 12. (11 (1):31 8. (1):31 8. 120 (9):1045 53. (9):1045 53. 18 (11 Suppl):S143 8. Suppl):S143 8. (11 124
20 BMJ Open http://bmjopen.bmj.com/ (2):131 51. (2):131 51. (1 Suppl):I275 81. Suppl):I275 81. (1 114 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml (11 Suppl):S60 5. Suppl):S60 5. (11 For peer review only (5):265 72. (5):265 72. (7):776 88. (7):776 88. 92 31 122 2010; Circulation 1162007; Circulation surgery. bypass graft artery coronary undergoing inwhites infarction myocardial 125 2009; Hum Genet Statement. STROBE after cardiac surgery. Circulation 2006; surgery. cardiac after 2008; 2007; artery bypass surgery in replicated analyses. Circulation 2011; Circulation analyses. inreplicated bypass surgery artery 10. Collard CD, Shernan SK, Fox AA, et al. The MBL2 'LYQA secretor' haplotype is an independent predictor of postoperative of independent predictor is an haplotype secretor' 'LYQA The MBL2 al. et FoxAA, SK, Shernan CD, Collard 10. 9. Muehlschlegel JD, Liu KY, Perry TE, et al. Chromosome 9p21 variant predicts mortality after coronary artery bypass graft surgery. bypass surgery. graft artery coronary after mortality predicts variant 9p21 Chromosome TE, al. et Perry JD, KY, Liu Muehlschlegel 9. 11. Podgoreanu MV, White WD, Morris RW, et al. Inflammatory gene polymorphisms and risk of postoperative myocardial infarction infarction myocardial postoperative of risk and polymorphisms gene Inflammatory RW, al. et White Morris WD, MV, Podgoreanu 11. 17. Skol AD, Scott LJ, Abecasis GR, et al. Optimal designs for two stage genome wide association studies. Genet Epidemiol Epidemiol Genet studies. association two stage genome wide for Optimal al. designs et GR, Abecasis Scott Skol AD, 17. LJ, 16. Little J, Higgins JP, Ioannidis JP, et al. Strengthening the reporting of genetic association studies (STREGA): an extension of the the of extension an (STREGA): studies association of genetic the reporting Strengthening al. JP, et JP, J, Ioannidis Higgins 16. Little 14. Luo L, Peng G, Zhu Y, et al. Genome wide gene and pathway analysis. Eur J Hum Genet 2010; J Genet analysis. Eur Hum pathway and gene Genome wide G, al. Peng Zhu et Y, L, 14. Luo 12. Lobato RL, White WD, Mathew JP, et al. Thrombomodulin gene variants are associated with increased mortality after coronary coronary after mortality increased with associated variants are gene JP, Thrombomodulin al. et White RL, Mathew WD, 12. Lobato 13. Khoriaty R, Vasievich MP, Ginsburg D. The COPII pathway and hematologic disease. Blood 2012; disease. hematologic Blood and pathway The D. COPII MP, Ginsburg R, Vasievich Khoriaty 13. 15. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by genome wide association. Genomics Genomics association. by assessed genome wide diseases of common seven analysis Topol Pathway NJ. A, EJ,Schork Torkamani 15. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (16):2020 35. (16):2020 35. 126 :57 67. :57 67. 47
21 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. (3):479 92. (3):479 92. (8):904 9. (8):904 9. 38 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (5):1644 53. (5):1644 53. 81 (19):2446 54. (19):2446 54. (7262):399 401. (7262):399 401. (4):e61114. (4):e61114. 8 294 461 development and human diseases. Int J Biochem 2014; Biol J Cell Biochem Int human diseases. and development 2013; studies. Nat Genet 2006; Genet Nat studies. of vein graft failure following coronary artery bypass graft surgery: PREVENT IV: a randomized controlled trial. Jama trial. randomizedcontrolled a PREVENT IV: surgery: bypassgraft artery coronary following failure graft vein of 2005; Immun 2013; Immun 191 J 2010; Biol Cell capacity. 2009; 25. Baines AC, Adams EJ, Zhang B, et al. Disruption of the Sec24d gene results in early embryonic lethality in the mouse. PLoS mouse. inthe One lethality embryonic results inearly Sec24d gene the of Disruption al. B, et Adams EJ, AC, Zhang Baines 25. 18. Alexander JH, Hafley G, Harrington RA, et al. Efficacy and safety of edifoligide, an E2F transcription factor decoy, for prevention prevention for decoy, transcription E2F factor an of safety edifoligide, and et Efficacy RA, al. Harrington G, JH, Hafley Alexander 18. 22. Fox RM, Hanlon CD, Andrew DJ. The CrebA/Creb3 like transcription factors are major and direct regulators of secretory secretory of regulators direct major and are factors transcription CrebA/Creb3 like DJ. The Andrew CD, FoxHanlon RM, 22. 19. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. Circulation 2012; Circulation infarction. myocardial of definition universal Third al. et AS, AlpertJS, Jaffe K, Thygesen 19. 24. Unlu G, Levic DS, Melville DB, et al. Trafficking mechanisms of extracellular matrix macromolecules: insights from vertebrate vertebrate from insights matrix macromolecules: extracellular of mechanisms Trafficking al. et Melville DS, DB, G, Unlu Levic 24. 20. Ge D, Fellay J, Thompson AJ, et al. Genetic variation in IL28B predicts hepatitis C treatment induced viral clearance. Nature clearance. viral predicts C treatment induced hepatitis in IL28B variation J, Genetic AJ, Fellay al. et D, Thompson Ge 20. 21. Price AL, Patterson NJ, Plenge RM, et al. Principal components analysis corrects for stratification in genome wide association ingenome wide for stratification analysis corrects components Principal al. et RM, Patterson NJ, AL, Plenge Price 21. 23. Ahyi AN, Quinton LJ, Jones MR, et al. Roles of STAT3 in protein secretion pathways during the acute phase response. response. Infect the acute phase during pathways secretion protein in STAT3 of Roles al. Jones et MR, AN, Quinton LJ, Ahyi 23. Page 21 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 22 of 51 (1):15 24. (1):15 24. 98 (3):407 15. (3):407 15. 29
(5):708 16. (5):708 16. 22 29 BMJ Open http://bmjopen.bmj.com/ (18):e211 2; author reply e11 2. reply author (18):e211 2; (2):469 77. (2):469 77. 33 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (2):229 37. (2):229 37. (10):3549 60. (10):3549 60. 45 22 (2):184 95. (2):184 95. 74 model of brief coronary occlusions. Cardiovasc Res 1995; Cardiovasc occlusions. coronary brief of model syndromes: comparison with markers of systemic inflammation, platelet activation, and myocardial necrosis. J Coll Am necrosis. myocardial and activation, platelet systemic inflammation, of markers with comparison syndromes: 2005; Cardiol 109 Circulation 2004; syndromes. coronary occlusion. Cardiovasc Res 1997; Cardiovasc occlusion. 2007; Faseb J 2008; Faseb after microembolisation in porcine heart. Cardiovasc Res 1995; Res Cardiovasc heart. porcine in microembolisation after 30. Conti E, Andreotti F, Zuppi C. Pregnancy associated plasma protein a as predictor of outcome in patients with suspected acute acute suspected with patients inoutcome of as protein predictor a plasma C. Pregnancy associated Zuppi F, Andreotti Conti E, 30. 28. Vogt AM, Htun P, Kluge A, et al. Insulin like growth factor II delays myocardial infarction in experimental coronary artery artery coronary inexperimental infarction delays myocardial factor II growth al. et P, Insulin like Kluge A, Htun AM, Vogt 28. 33. Bujak M, Frangogiannis NG. The role of TGF beta signaling in myocardial infarction and cardiac remodeling. Cardiovasc Res Res Cardiovasc remodeling. and cardiac infarction inmyocardial signaling of TGF beta role The NG. Frangogiannis M, Bujak 33. 27. Kluge A, Zimmermann R, Munkel B, et al. Insulin like growth factor II is an experimental stress inducible gene in a porcine porcine ina gene stress inducible is an factor experimental II growth al. et Insulin like B, R, Munkel A, Zimmermann Kluge 27. 31. Granger A, Abdullah I, Huebner F, et al. Histone deacetylase inhibition reduces myocardial ischemia reperfusion injury in mice. inmice. injury ischemia reperfusion myocardial inhibition reduces deacetylase Histone al. F, et Huebner Abdullah I, A, Granger 31. 29. Heeschen C, Dimmeler S, Hamm CW, et al. Pregnancy associated plasma protein A levels in patients with acute coronary coronary acute with levels patients in plasma protein A Pregnancy associated CW, al. et S, Hamm C, Dimmeler Heeschen 29. 32. Backs J, Olson EN. Control of cardiac growth by histone acetylation/deacetylation. Circ Res 2006; Res Circ acetylation/deacetylation. histone by growth cardiac of Control J, EN. Olson Backs 32. 26. Kluge A, Zimmermann R, Munkel B, et al. Insulin like growth factor I is involved in inflammation linked angiogenic processes processes angiogenic linked factor ininflammation is involved I growth al. et Insulin like B, R, Munkel A, Zimmermann Kluge 26. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (p) for all single nucleotide polymorphisms nucleotide single all for (p) 10 (3):351 60. (3):351 60. ), but 521 SNPs met the prespecified discovery threshold threshold discovery prespecified the SNPs met 521 but ), 98 −8
, rs2044061 on chromosome 8). chromosome on rs2044061 , 6 23 <9.09×10 p BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (6):653 61. (6):653 61. 57 troponin allows more accurate diagnosis of periprocedural necrosis and infarction after coronary intervention. J intervention. Coll Am coronary after infarction and necrosis periprocedural of diagnosis accurate more allows troponin 2011; Cardiol protects the heart from ischemia/reperfusion injury. Circ Res 2006; Res Circ injury. ischemia/reperfusion from heart the protects
(SNPs). None of the SNPs reached genome wide significance ( significance SNPs the genome wide of reached None (SNPs). 34. Kempf T, Eden M, Strelau J, et al. The transforming growth factor beta superfamily member growth differentiation factor 15 factor 15 growth differentiation member superfamily factor beta growth transforming Strelau J,M, The al. Eden et T, Kempf 34. LEGENDS FIGURE x axis The analysis. I ininfarction Stage myocardial perioperative with association of1. plot genome wide Manhattan Figure the –log shows chromosome), by y axis order (colored in physical the genome represents p< 0.001 for inclusion in Stage II analyses (minimum p = 2.76x10 = p (minimum analyses in inclusion II Stage for 0.001 p<
35. Lim CC, van Gaal WJ, Testa L, et al. With the "universal definition," measurement of creatine kinase myocardial band rather than than band rather kinase myocardial creatine of measurement definition," With "universal al. the WJ, et Gaal CC, van L, Testa 35. Lim Page 23 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 24 of 51 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only
Figure 1 Figure 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 51 BMJ Open
1 SUPPLEMENTAL MATERIAL BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 Table of Contents 6 7 8 PEGASUS Investigative Team Members Page 2 9 10 Supplemental Methods Page 3 11 12 A.1. Quality control of data flow Page 3 13 14 A.2. Controlling for population stratification Page 4 15 For peer review only 16 A.3. Haplotype analysis Page 6 17 18 A.4. GWAS pathway analysis Page 6 19 20 21 Supplemental Results Page 7 22 23 B.1. Patient characteristics Page 7 24 25 B.2. Single locus GWAS analysis Page 8 26 27 B.3. Haplotype analysis Page 8 28 29 B.4. Chromosome 9p21 locus analysis Page 14 30 31 B.5. GWAS pathway analysis results Page 15 32 33 34 B.6. Power calculation Page 21 http://bmjopen.bmj.com/ 35 36 B.7. Details of the top 521 SNPs identified in stage I GWAS analyses Page 22 37 38 Supplemental References Page 25 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 1
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 51
1 Perioperative Genetics and Safety Outcomes Study (PEGASUS) Investigative Team Members BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Allen AS, Davis RD, Funk B, Gaca JG, Ginsburg GS, Glower DD, Goldstein DB, Grichnik KP, Hall RL, 4 5 Hauser E, Jones R, Kertai MD, Laskowitz DT, Li YJ, Lodge AJ, Mathew JP, Milano CA, Moretti EW, 6 7 8 Newman MF, Phillips Bute B, Podgoreanu MV, Smith MP, Smith PK, Stafford Smith M, Swaminathan 9 10 M, Welsby IJ, White WD, Willard HF 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 2
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 51 BMJ Open
1 A. SUPPLEMENTAL METHODS BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 A.1. Quality control of data flow 4 5 The following quality control steps were taken to make sure that genotypes were correctly called. 6 7 8 • Infinium BeadStudio raw data analysis: Any sample that had a very low intensity or a very low 9 10 call rate using the standard Illumina cluster (< 95%) was deleted. All SNPs that had a call 11 12 frequency below 100% were then reclustered. Any sample that was below a 98% call rate after 13 14 reclustering was deleted. Next, all SNPs that had a call frequency below 99% were deleted. Any 15 For peer review only 16 SNPs for which >1% of samples were not called or were ambiguously called were deleted. A 17 18 total of 35 samples (13 in GeneMagic and 22 in PEGASUS cohort) were deleted during this 19 20 21 procedure. 22 23 • Minor allele frequency (MAF) check for data handling accuracy: This step performs a basic 24 25 check of the data accuracy on the data flow pipeline from the output of the Illumina genotyping 26 27 facility to the analytical process. We checked the MAF report from PLINK software against the 28 29 original locus report generated by the genotyping facility, and ensured that the 2 matched exactly. 30 31 32 • Gender specification: This step compares the observed genotypes on chromosomes X and Y to 33 34 the gender specification obtained from the phenotype database. Samples with X chromosomal http://bmjopen.bmj.com/ 35 36 heterozygosity that was inconsistent with reported gender were individually inspected against the 37 38 original data source. A total of 11 samples could not be reconciled and were excluded at this step 39 40 (8 in GeneMagic and 3 in PEGASUS cohort) 41
42 on October 1, 2021 by guest. Protected copyright. • Cryptic relatedness: We estimated the sharing of genetic information between cohort participants 43 44 45 by calculating identity by descent (IBD) using PLINK software. All pairs with DNA samples 46 47 showing ≥ 0.125 estimated proportion of alleles IBD were inspected, and one sample from each 48 49 pair was excluded from further analyses. A total of 9 samples (all in PEGASUS cohort) were 50 51 excluded at this step. 52 53 • Genotype missing: This step assesses whether the genotype missing is skewed toward cases or 54 55 56 controls, which could give rise to spurious association p values. We used PLINK software to 57 58 perform this check on the top SNPs discussed in the paper. 59 60 3
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 51
1 • Low MAF: We removed all SNPs with MAF < 0.01 in order to control for error in the estimation BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 of asymptotic p values, as alleles with such a low MAF would have no chance of approaching 4 5 significance based on our sample size and perioperative MI incidence. 6 7 8 • Hardy Weinberg Equilibrium (HWE): We used PLINK software to identify any observed 9 6 10 genotypes that deviated from HWE. All markers that significantly deviated from HWE (p < 10 ) 11 12 were excluded. 13 14 Analyses were performed using PLINK 1.07, SAS/Genetics version 9.2, and R version 2.15.1. 15 For peer review only 16 17 18 19 A.2. Controlling for population stratification 20 1 21 We used a modified EIGENSTRAT method to control for population stratification. This method derives 22 23 the principal components of the correlations among the gene variants, and corrects for those correlations 24 25 in the association tests. Population structure was investigated using the EigenSoft program. All 15 26 27 principal components (PCs) were computed for both datasets. Plots between pairwise PCs, particularly 28 29 PC1 vs PC2, were generated to determine whether any obvious outliers deviated from the main cluster, 30 31 and hence, had to be excluded from subsequent analyses (Supplemental Figure S1). 32 33 34 http://bmjopen.bmj.com/ 35 36 37 Supplemental Figure S1. Comparison of PC1 38 39 vs PC2 reveals potential population 40 41 stratification in the discovery (Gene Magic) 42 on October 1, 2021 by guest. Protected copyright. 43 dataset. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 4
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 51 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 In addition, we performed multivariate regression analyses using a strategy to keep all PCs up to 5 6 the last PC with p < 0.05 in each step, to determine the number of PCs to be used for correction in the 7 8 final analysis. For instance, starting with 15 PCs in the model, if PC(i) is the last PC with p < 0.05, we 9 10 11 included PC(1) to PC(i) in the next multivariate model, and then repeated the process until the last PC 12 13 remained nominally significant. Using this iterative multivariate analysis, we found that the top 10 PCs 14 15 were a reasonableFor set for the peerGene Magic dataset review (p = 0.032), and onlywere subsequently re estimated and 16 17 used for stage II analysis (Supplemental Table S1). This final set of 10 PCs was then used as covariates 18 19 to adjust for ancestry, along with clinical variables, in multivariate logistic regression analysis of 20 21 perioperative MI. 22 23 24 Supplemental Table S1: Principal component analysis of the study populations 25 26 27 ANALYSIS OF MAXIMUM LIKELIHOOD ESTIMATES 28 29 DISCOVERY DATSET REPLICATION DATASET 30 Parameter DF Estimate Standard Wald chi- P-value DF Estimate Standard Wald chi- P-value 31 error square test error square test 32 33 Intercept 1 2.635 0.127 434.325 <0.0001 1 2.153 0.076 798.277 <0.0001 http://bmjopen.bmj.com/ 34 Principal component 1 1 31.474 14.214 4.903 0.027 1 18.429 8.413 4.798 0.029 35 36 Principal component 2 1 5.486 3.826 2.056 0.152 1 4.009 3.338 1.443 0.230 37 Principal component 3 1 7.575 3.497 4.691 0.03 1 3.239 3.230 1.006 0.316 38 39 Principal component 4 1 0.431 3.479 0.015 0.901 1 1.266 3.119 0.165 0.685 40 Principal component 5 1 18.941 9.623 3.874 0.049 1 1.642 4.828 0.116 0.734 41 Principal component 6 1 0.519 7.944 0.004 0.948 1 15.494 7.678 4.073 0.044 42 on October 1, 2021 by guest. Protected copyright. 43 Principal component 7 1 2.231 10.001 0.049 0.824 1 12.985 6.873 3.569 0.059 44 Principal component 8 1 10.793 9.043 1.424 0.233 1 2.244 4.825 0.216 0.642 45 46 Principal component 9 1 29.876 11.661 6.564 0.01 1 8.120 5.366 2.289 0.130 47 Principal component 10 1 22.775 10.593 4.623 0.032 1 7.487 4.650 2.592 0.107 48 49 DF, degrees of freedom 50 51 52 53 54 55 56 57 58 59 60 5
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 51
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 A.3. Haplotype analysis 4 5 For the final candidate genes prioritized based on the significance of SNP association tests, we first 6 7 identified the linkage disequilibrium (LD) blocks harboring the significant SNPs. Within each LD block, 8 9 10 we performed haplotype association analysis using sliding windows of 3 markers across the region. This 11 12 was conducted for the replication dataset, by constructing multivariable logistic regression models 13 14 adjusting for the same clinical covariates and PCs used for single marker analyses. 15 For peer review only 16 17 18 A.4. GWAS pathway analysis 19 20 SNPs with a covariate adjusted association p value < 0.01 in stage I GWAS were subjected to functional 21 22 23 ontology enrichment analyses using tools implemented in the MetaCore software suite (GeneGO, St. 24 25 Joseph, MI). Enrichment analysis begins by assigning SNPs to genes, and matching these geneIDs with 26 27 geneIDs in functional ontologies in highly curated canonical pathway maps in MetaCore. The aggregation 28 29 of GWAS associations for the genes in tested pathways, ie, the significance of enrichment, is computed as 30 31 the p value of a hypergeometric model.2 Canonical pathways enriched in stage I and stage II analyses 32 33
were further compared; and the probability of a random intersection between the set of genes in common http://bmjopen.bmj.com/ 34 35 36 in the 2 cohorts, and functional ontology entities was estimated as p value of a hypergeometric 37 38 distribution in MetaCore. 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 6
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 51 BMJ Open
1 B. SUPPLEMENTAL RESULTS BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 B.1. Patient characteristics 4 5 Supplemental Table S2. Clinical characteristics of the study populations 6 7 STAGE I ANALYSIS (N=1433) STAGE II ANALYSIS (N=2055) 8 No PMI, PMI No PMI PMI 9 Characteristics OR (95%CI) P-value* OR (95%CI) P-value* 10 (n=1321) (n=112) (n=1830) (n=225) 11 Extracardiac arteriopathy 251 (19.0) 20 (17.9) 0.93 (0.56 1.53) 0.77 754 (41.2) 128 (56.9) 1.88 (1.42 2.49) <0.0001 12 Recent myocardial infarction 256 (19.4) 23 (20.5) 1.08 (0.67 1.74) 0.77 393 (21.5) 39 (17.3) 0.77 (0.53 1.10) 0.15 13 Procedure other than CABG only 126 (9.5) 23 (20.5) 2.45 (1.50 4.02) 0.0004 153 (8.4) 51 (22.7) 3.21 (2.26 4.58) <0.0001 14 Year of surgery by year 15 1997 For peer review 26 (1.4)only 3 (1.3) 1.0 16 1998 70 (3.8) 8 (3.6) 0.99 (0.24 4.02) 0.98 17 1999 87 (4.8) 3 (1.3) 0.30 (0.06 1.57) 0.15 18 2000 77 (4.2) 14 (6.2) 1.58 (0.42 5.92) 0.50 19 2001 49 (2.7) 12 (5.3) 2.12 (0.55 8.20) 0.28 20 21 2002 146 (11.1) 9 (8.0) 1.0 188 (10.3) 20 (8.9) 0.92 (0.26 3.32) 0.90 22 2003 1175 (88.9) 103 (92.0) 1.42 (0.70 2.87) 0.33 1189 (65.0) 109 (48.4) 0.80 (0.24 2.67) 0.71 23 2004 32 (1.8) 8 (3.6) 2.17 (0.52 9.0) 0.29 24 2005 80 (4.4) 36 (16.0) 3.90 (1.11 13.72) 0.03 25 2006 32 (1.8) 12 (5.3) 3.25 (0.83 12.75) 0.09 26 Number of diseased vessels 27 0 67 (5.1) 6 (5.3) 1.0 69 (3.8) 6 (2.7) 1.0 28 1 259 (19.6) 29 (25.9) 1.25 (0.50 3.14) 0.63 276 (15.0) 32 (14.2) 1.33 (0.54 3.32) 0.54 29 2 519 (39.3) 43 (38.4) 0.93 (0.38 2.26) 0.86 595 (32.5) 52 (23.1) 1.01 (0.42 2.43) 0.99 30 3 476 (36.0) 34 (30.4) 0.80 (0.32 1.97) 0.62 885 (48.4) 130 (57.8) 1.69 (0.72 3.97) 0.22 31 4 5 (0.3) 5 (2.2) 11.5 (2.58 51.24) 0.0014 32 Aortic cross clamp time, min 51.2+36.8 71.9+32.4 1.015 (1.01 1.02) <0.0001 55.0+35.2 80.5+36.7 1.019 (1.015 1.023) <0.0001 33 34 Continuous variables are presented as means+standard deviation, and categorical variables as percent frequencies. CABG, coronary artery byp http://bmjopen.bmj.com/ 35 Descriptive statistics of clinical variables are presented as count (percent frequency) for categorical 36 37 variables and mean ± SD for continuous variables. PMI, postoperative myocardial infarction; CABG, 38 39 coronary artery bypass surgery; extracardiac arteriopathy, defined as any one or more of the following: 40 41
42 claudication, carotid occlusion or > 50% stenosis, and/or previous or planned intervention on the on October 1, 2021 by guest. Protected copyright. 43 44 abdominal aorta, limb arteries or carotids; OR (95%CI), univariate odds ratio (95% confidence interval). 45 46 *Comparisons made using the t test or Wald chi square test, as appropriate. Clinical variables were 47 48 included in the final multivariate logistic regression analysis if significant (p < 0.05) in univariate 49 50 analyses. 51 52 53 54 55 56 57 58 59 60 7
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 51
1 B.2. Single-locus GWAS analysis - population structure adjustment BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8 Supplemental Figure S2. Quantile Quantile (Q Q) 9 10 plot representation of the GWAS association results 11 12 in the discovery dataset 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 Distribution of expected (under the null hypothesis) vs observed p values for single nucleotide 26 27 polymorphisms that are associated with perioperative myocardial infarction, based on a whole genome 28 29 wide analysis. The red diagonal line represents the line obtained if the observed distribution did not 30 31 deviate from the expected distribution. All p values were corrected for the inflation factor (λ = 1.0073). 32 33
http://bmjopen.bmj.com/ 34 35 36 B.3. Haplotype analysis of genes harboring top scoring SNPs 37 38 Supplemental Figure S3. Linkage disequilibrium (LD) structures of [A] PAPPA2 (1q23 q25 39 40 chromosome region), [B] SEC24D (4q26 chromosome region), and [C] HDAC4 (2q37.3 chromosome 41 42 region), and displayed as pairwise correlation plots using HapMap database (CEU European ancestry) on October 1, 2021 by guest. Protected copyright. 43 44 (Haploview). Regions of LD are shaded in bright red (strong LD) and lighter for moderate or weak LD. 45 46 47 The physical location of the individual SNPs associated with perioperative myocardial infarction in 2 48 49 stage analyses is circled in red. True haplotype blocks in the population are marked with black lines in the 50 51 correlation plot. 52 53 54 55 56 57 58 59 60 8
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 51 BMJ Open
1 A. PAPPA2 gene BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8 9 10 11
12 13 14 15 For peer review only 16 17 18 19 20 21 22 B. SEC24D gene 23 24 25 26 27 28
29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 1, 2021 by guest. Protected copyright. 43 C. HDAC4 gene 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 34 of 51 0.759 0.759 1.58 12.2 0.000477 0.193 0.193 0.515 18.8 1.45E-05 0.655 0.655 0.179 1.43 1.08 9.56 0.292 0.00199 0.589 0.197 0.755 0.5 1.62 0.245 0.675 20.3 0.616 13.4 6.47E-06 1.49 0.244 0.000248 13.4 0.676 0.618 11.4 0.000256 1.49 0.00075 13.2 11.3 0.000274 0.000796 0.166 0.166 0.464 19.5 1.01E-05 0.576 0.576 1.29 5.58 0.0182 0.344 0.344 0.698 9.56 0.00198 0.248 0.248 1.19 0.589 1.25 2.17 0.241 0.631 0.141 4.47 12.2 0.0345 0.000477 0.175 0.175 0.474 20.8 4.98E-06 0.522 1.17 2.38 0.123 0.237 0.237 1.23 3.16 0.0754 0.578 0.578 1.29 0.175 0.474 5.55 20.8 0.0184 4.98E-06 0.237 0.237 1.23 3.16 0.0754 0.0485 0.0485 1.28 1.24 0.266 0.0485 0.0485 1.28 0.0795 1.24 1.02 0.266 0.0801 0.0109 1.02 0.917 0.00645 0.936 0.0786 0.0786 1.22 1.16 0.282 AA AA GA GA GA GA AG AG AG AG GG GG GG GG 10 BMJ Open http://bmjopen.bmj.com/ PAPPA2 and HDAC4 HDAC4 and PAPPA2 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only PAPPA2 PAPPA2 PAPPA2 PAPPA2 176562608 176562796 PAPPA2 176562608 RS10489477 176562796 RS1395460 PAPPA2 176562796 RS10489477 176565307 RS1395460 PAPPA2 176562796 RS1395460 176565307 PAPPA2 176562796 RS909941 RS1395460 GG 176565307 PAPPA2 176565307 RS909941 RS1395460 GA 176566350 PAPPA2 176565307 RS909941 RS909941 176566350 PAPPA2 176565307 RS10489478 RS909941 176566350 PAPPA2 176566350 RS10489478 RS909941 176571431 PAPPA2 176566350 RS10489478 RS10489478 176571431 GA PAPPA2 RS240104 176566350 RS10489478 176571431 AG PAPPA2 RS240104 176571431 RS10489478 176571578 GG PAPPA2 RS240104 176571431 RS240104 176571578 PAPPA2 176571431 RS10798463 RS240104 176571578 PAPPA2 176571578 RS10798463 RS240104 176589732 PAPPA2 176571578 RS10798463 RS10798463 176589732 RS12139318 GG PAPPA2 176571578 RS10798463 176589732 RS12139318 AA PAPPA2 176589732 RS10798463 176591291 RS12139318 GA PAPPA2 176589732 RS12139318 GC 176591291 RS3795320 PAPPA2 176589732 RS12139318 AC 176591291 RS3795320 PAPPA2 176591291 RS12139318 GA 176593524 RS3795320 PAPPA2 176591291 RS3795320 CC 176593524 PAPPA2 RS12565327 176591291 RS3795320 CA 176593524 PAPPA2 RS12565327 176593524 RS3795320 AC 176603840 RS12565327 176593524 RS12565327 CG 176603840 RS7529392 176593524 RS12565327 AG 176603840 RS7529392 RS12565327 AA RS7529392 GA GG AG PAPPA2 PAPPA2 176562608 176562796 RS10489477 RS1395460 GENESYMBOL BP1 BP2 SNP1 SNP2 HAPLOTYPE F OR STAT P
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 CHR Results of the haplotype analyses for analyses thehaplotype of Results S3. Table Supplemental 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 0.176 0.176 0.461 22.1 2.65E-06 0.176 0.461 22.1 2.60E-06 0.489 0.489 1.29 0.663 5.98 1.42 0.0145 9.13 0.00252 0.756 0.756 1.62 0.489 13.3 1.29 0.351 0.000269 0.665 5.98 12.2 0.0145 0.000484 0.161 1.17 1.39 0.239 0.176 0.176 0.46 22.1 2.56E-06 0.336 0.336 1.16 0.161 1.85 1.17 0.174 1.39 0.239 0.177 0.177 0.756 0.46 1.62 22.2 13.3 2.52E-06 0.000264 0.227 0.721 5.97 0.0146 0.706 0.706 1.23 3.05 0.0806 0.245 0.245 0.625 12.7 0.000371 0.591 0.591 1.28 5.11 0.0238 0.164 0.164 1.18 1.44 0.231 0.232 0.232 1.23 3.09 0.0787 0.591 0.591 1.28 5.11 0.769 0.0238 0.832 2.43 0.119 0.176 0.176 0.461 22.1 2.59E-06 0.164 0.164 1.18 1.42 0.234 0.061 0.061 2.19 18.6 1.58E-05 0.061 0.061 2.18 18.4 1.79E-05 0.276 0.276 0.962 0.102 0.749 0.663 0.663 0.804 3.83 0.0503 0.0674 0.0674 1.23 1.13 0.288 0.0675 0.0675 1.23 1.13 0.288 0.0615 1.17 0.537 0.464 0.0665 0.0665 1.2 0.831 0.362 AA AA AA AA AA AA AA AA GA GA AG AG GC GC GG GG 11 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only HDAC4 HDAC4 HDAC4 HDAC4 240201797 240208154 240201797 RS13414769 240208154 RS2048765 240208154 RS13414769 240216021 RS2048765 RS2048765 RS6737742 GA GG HDAC4 HDAC4 240201797 240208154 RS13414769 RS2048765 GG PAPPA2 PAPPA2 PAPPA2 PAPPA2 176603840 176604564 176603840 RS7529392 176604564 PAPPA2 RS12133739 176603840 RS7529392 176604564 PAPPA2 RS12133739 RS7529392 RS12133739 176604564 AA 176630416 PAPPA2 176604564 RS12133739 GA 176630416 RS10913237 PAPPA2 RS12133739 AG RS10913237 PAPPA2 176630416 176646383 PAPPA2 176630416 RS10913237 AG 176646383 RS10798473 PAPPA2 176646383 RS10913237 GG 176647000 RS10798473 176646383 RS10798473 176647000 RS17546000 PAPPA2 176646383 RS10798473 GA 176647000 RS17546000 PAPPA2 RS10798473 GG RS17546000 176647000 GA 176649181 PAPPA2 176647000 RS17546000 AA 176649181 RS10454444 PAPPA2 RS17546000 GG RS10454444 PAPPA2 176649181 176660247 PAPPA2 176649181 RS10454444 AA 176660247 PAPPA2 RS726252 176660247 RS10454444 GA 176667810 PAPPA2 RS726252 176660247 RS726252 176667810 PAPPA2 176660247 RS726252 176667810 RS791031 PAPPA2 176667810 RS726252 176680137 RS791031 PAPPA2 176667810 RS791031 176680137 RS791031 PAPPA2 176667810 RS12138021 RS791031 176680137 PAPPA2 176680137 RS12138021 RS791031 176684432 176680137 RS12138021 RS12138021 176684432 AA RS4507975 176680137 RS12138021 176684432 CA RS4507975 RS12138021 AG RS4507975 AG GG PAPPA2 PAPPA2 176604564 176630416 PAPPA2 RS12133739 RS10913237 176630416 176646383 RS10913237 AA RS10798473 PAPPA2 AG 176647000 176649181 PAPPA2 RS17546000 RS10454444 176649181 176660247 RS10454444 AG RS726252
2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1
Page 35 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 36 of 51 0.13 0.13 0.905 0.388 0.533 0.52 0.52 0.965 0.112 0.737 0.73 0.73 0.75 6.47 0.011 0.51 0.51 0.8 4.6 0.032 0.546 0.546 1.07 0.387 0.534 0.382 0.382 0.986 0.0174 0.895 0.512 0.512 0.945 0.286 0.593 0.382 0.382 0.986 0.0174 0.895 0.488 0.488 1.06 0.301 0.583 0.723 1.09 0.482 0.487 0.524 0.524 0.968 0.0946 0.758 0.415 0.415 0.819 3.31 0.0689 0.185 0.185 1.01 0.00304 0.956 0.208 1.02 0.0282 0.867 0.423 0.423 0.84 2.58 0.108 0.718 0.718 0.773 5.06 0.0244 0.224 0.224 1 0.00029 0.986 0.414 0.812 3.61 0.0575 0.221 0.221 0.978 0.0301 0.862 0.266 0.266 1.32 5.92 0.015 0.594 0.594 1.01 0.00252 0.96 0.221 0.221 0.986 0.0128 0.91 0.221 0.221 0.986 0.012 0.913 0.182 0.182 1.01 0.00393 0.95 0.594 0.594 0.995 0.0024 0.961 0.0721 0.0721 0.827 0.819 0.365 0.0645 0.0645 2.13 17.9 2.32E-05 0.0721 0.0721 0.827 0.823 0.364 0.0615 0.0615 2.14 17.7 0.000026 0.0649 0.0649 2.12 17.8 2.43E-05 0.0614 0.0614 2.21 19.6 9.58E-06 AA AA AA AA AA AC AC GA GA GA GA GA GA GA AG AG AG AG AG AG GC GC GG GG GG GG GG GG 12 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only HDAC4 HDAC4 HDAC4 HDAC4 240208154 240216021 HDAC4 240208154 RS2048765 240216021 HDAC4 240216021 RS6737742 RS2048765 240217116 HDAC4 240216021 RS6737742 RS6737742 240217116 HDAC4 RS11883623 240216021 RS6737742 240217116 HDAC4 AG RS11883623 240217116 RS6737742 240217522 HDAC4 RS11883623 240217116 RS11883623 AG 240217522 HDAC4 RS908262 240217116 RS11883623 GG 240217522 HDAC4 RS908262 240217522 RS11883623 GA 240228263 RS908262 240217522 RS908262 240228263 HDAC4 240217522 RS908262 RS6543523 240228263 HDAC4 RS908262 RS6543523 HDAC4 240228263 RS6543523 240234901 HDAC4 240228263 CG RS6543523 240234901 HDAC4 RS10200850 240234901 CA RS6543523 240241255 HDAC4 RS10200850 240234901 RS10200850 240241255 HDAC4 RS925738 240234901 RS10200850 AG 240241255 HDAC4 RS925738 240241255 RS10200850 GG 240250456 HDAC4 RS925738 240241255 RS925738 240250456 HDAC4 240241255 RS925738 RS7594398 240250456 HDAC4 240250456 RS925738 RS7594398 240257958 HDAC4 240250456 RS7594398 RS7594398 240257958 HDAC4 240250456 GA RS1399629 RS7594398 240257958 HDAC4 240257958 AG RS1399629 RS7594398 240262287 HDAC4 240257958 GG RS1399629 RS1399629 240262287 HDAC4 240257958 RS908265 RS1399629 240262287 HDAC4 GA 240262287 RS908265 RS1399629 240263103 HDAC4 GG 240262287 RS908265 RS908265 240263103 HDAC4 240262287 RS908265 240263103 RS908263 240263103 RS908265 240263584 RS908263 240263103 RS908263 240263584 RS908263 RS10211084 RS908263 RS10211084 AG GG HDAC4 HDAC4 240228263 240234901 RS6543523 RS10200850 GA
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 0.17 0.17 1.02 0.0248 0.875 0.52 0.52 1.27 5.11 0.0238 0.31 0.31 0.747 6.34 0.0118 0.485 0.485 1.29 6.18 0.0129 0.205 0.205 0.978 0.0302 0.862 0.205 0.205 0.978 0.0302 0.862 0.309 0.309 0.748 6.28 0.0122 13 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only HDAC4 HDAC4 HDAC4 HDAC4 240263103 240263584 HDAC4 240263584 RS908263 240265617 HDAC4 240263584 RS10211084 RS10211084 240265617 HDAC4 RS6706275 240263584 RS10211084 240265617 HDAC4 RS6706275 240265617 RS10211084 240275570 GA RS6706275 240265617 RS6706275 GA 240275570 RS12988669 240265617 RS6706275 GG 240275570 RS12988669 RS6706275 AG RS12988669 AA GA GG
2 2 2 2 2 2 2
Page 37 of 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open Page 38 of 51
1 B.4. Chromosome 9p21 locus and perioperative MI BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Supplemental Figure S4. Single nucleotide polymorphisms at 9p21 locus previously associated with 4 5 incident myocardial infarction3 and mortality4 after CABG. 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 The dashed horizontal line indicates the nominal significance threshold, p = 0.05. The green dots 37 38 39 represent single nucleotide polymorphisms (base pair) at the 9p21 locus previously identified as 40 41 predictors of perioperative MI: rs1547705 (22082375, intron); rs1333040 (22083404, intron); rs4977574
42 on October 1, 2021 by guest. Protected copyright. 43 (22098574, intron); rs944797 (22115286, intron); rs2383207 (22115959, intron); rs1537375 (22116071, 44 45 intron). The red dot represents rs10116277 (22081397) variant, which was previously associated with 46 47 myocardial infarction in both non surgical and cardiac surgical cohorts. The dark blue point rs1381923 48 49 (22909438, unknown function) is the most significant variant in this plotted region. All p values are the 50 51 52 results of multivariate risk factor adjusted analyses. 53 54 55 56 57 58 59 60 14
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 51 BMJ Open
1 B.5. GWAS pathway analysis results BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Of the 534,350 SNPs included in analysis, 250,740 were recognized, and 66,893 were mapped to 19,163 4 5 genes within MetaCore. Stage I cohort analysis in MetaCore identified several canonical pathways after 6 7 8 functional enrichment analysis with SNPs that passed the p value and odds ratio filters (p < 0.01, 0.9 < 9 10 OR > 1.1). 11 12 Supplemental Figure S5. GenGo Canonical Pathway Maps histograms after enrichment analysis with 13 14 SNPs that passed the threshold and p value filter in GWAS stage I analyses. 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 The top 10 canonical maps for enriched SNPs are shown as histogram plots. Results are tabulated/sorted 50 51 in the histogram based on calculated –log(p values) of hypergeometric distribution statistics. 52 53 54 55 56 57 58 59 60 15
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 51
1 Supplemental Figure S6. The top scoring pathway map based on the enrichment distribution in GWAS BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 stage I analysis – Cell adhesion: extracellular matrix remodeling. 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 Top scoring SNPs based on GWAS stage I results are linked to genes and visualized on the maps as on October 1, 2021 by guest. Protected copyright. 43 44 thermometer like symbols. Up ward thermometers are red and indicate an odds ratio >1 for association 45 46 47 with perioperative MI, and down ward (blue) ones indicate an odds ratio for association < 1. Letters “B”, 48 49 and “C” indicate physical interactions, B, binding; and C, cleavage. Arrows with hexagons indicate 50 51 positive (green), negative (red), and unspecified (gray) effects. Arrows without hexagons indicate 52 53 technical link (gray). For further information on pathway elements please see: 54 55 http://pathwaymaps.com/pdf/MC_legend.pdf 56 57 58 59 60 16
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 51 BMJ Open
1 Top scoring pathways in stage I analyses were compared with enriched pathways in the stage II cohort. BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 The probability of a random intersection between a set of genes common to both cohorts with functional 4 5 ontology entities was estimated as p value of hypergeometric distribution using the Compare Experiments 6 7 8 Workflow in MetaCore. The top 4 scored canonical pathway maps (lowest p values), based on the 9 10 enrichment distribution, are presented below. 11 4 12 1. Immune response_NFAT signaling and leukocyte interaction (p = 7.2*10 ). Nuclear factors of 13 14 activated T cells (NFATs) is a family of calcium dependent transcription factors with pivotal roles in 15 For peer review only 16 regulating immune responses, which include interactions between antigen presenting cells and other 17 18 leukocytes, and expression of a variety of cytokines by coupling changes in intracellular calcium 19 20 21 concentration to gene expression. Variants in CD247, calcium channels, MHC class II, and NFAT that 22 23 map to this pathway, and their association with perioperative MI is presented below in Supplemental 24 25 Table S4. 26 27 28 SNPid Gene Symbol Gene Name OR p-value 29 30 rs2982480;rs2984800;rs2995054 CD247 CD247 molecule 2.16 0.00311 31 rs42051;rs7804449;rs1544462;rs1156325; calcium channel, voltage-dependent, alpha 2/delta 32 CACNA2D1 0.71 0.00368 33 rs2237526;rs2018982 subunit 1 http://bmjopen.bmj.com/ 34 rs1779244;rs2068357 CACNB2 calcium channel, voltage-dependent, beta 2 subunit 0.50 0.00162 35 major histocompatibility complex, class II, DP beta 2 rs3117016;rs9380342 HLA-DPB2 2.45 0.00795 36 (pseudogene) 37 38 rs3129882 HLA-DRA major histocompatibility complex, class II, DR alpha 1.49 0.00580 nuclear factor of activated T-cells, cytoplasmic, 39 rs1562724;rs1017860;rs8090864 NFATC1 1.45 0.00362 40 calcineurin-dependent 1 41 42 on October 1, 2021 by guest. Protected copyright. 43 2. Cell adhesion and extracellular matrix (ECM) remodeling (p = 1.1*10 3). An important pathway 44 45 46 involved in responses to myocardial injury, is comprised of matrix metalloproteinases (MMPs), their 47 48 endogenous tissue inhibitors (TIMPs), and various collagens. Furthermore, the plasminogen plasmin 49 50 system and its regulators (PAI 1) are implicated in proteolytic degradation of ECM. Finally, 2 categories 51 52 of growth factors are part of this canonical pathway – insulin like growth factors 1 and 2 (IGF 1 and IGF 53 54 2), and heparin binding EGF like growth factor (HB EGF) – and all are involved in regulating cell 55 56 growth, proliferation, and survival. IGF 1 and IGF 2 function by activating the IGF 1 receptor, but their 57 58 59 60 17
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 51
1 bioavailability is regulated by a family of insulin like growth factor binding proteins (IGFBP4)., HB EGF BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 activates the epidermal growth factor receptor (EGFR) and v erb a erythroblastic leukemia viral oncogene 4 5 homolog 4 (ErbB4), leading to cell proliferation, cell survival, and tissue remodeling (Supplemental 6 7 8 Table S5). 9 10 11 SNPid Gene Symbol Gene Name OR P-value 12 13 rs953386;rs2391823 COL4A1 collagen, type IV, alpha 1 1.76 0.00162 14 epidermal growth factor rs884904;rs845552 EGFR 1.67 0.00238 15 For peer reviewreceptor only 16 rs9288433;rs3817429;rs2371276;rs 1473636;rs7425448;rs10932384;rs6 v-erb-a erythroblastic 17 740117;rs12053361;rs17803152;rs7 ERBB4 leukemia viral oncogene 0.53 0.00613 18 594456;rs4672615;rs1851169;rs674 homolog 4 (avian) 19 7637 insulin-like growth factor 1 20 rs2033178 IGF1 0.50 0.00901 21 (somatomedin C) 22 rs8038056;rs8028620;rs2139924;rs insulin-like growth factor 1 IGF1R 1.90 0.00383 23 3784604 receptor 24 insulin-like growth factor 2 25 rs734351 IGF2 1.57 0.00159 (somatomedin A) 26 27 rs12454479;rs8086477;rs12454564 LAMA3 laminin, alpha 3 2.01 0.00165 28 matrix metallopeptidase 29 rs762052 MMP14 1.60 0.00863 14 (membrane-inserted) 30 31 serpin peptidase inhibitor, clade E (nexin, 32 rs1050813 SERPINE1 plasminogen activator 1.38 0.00874 33 inhibitor type 1), member 34 1 http://bmjopen.bmj.com/ 35 rs783147 PLG plasminogen 0.68 0.00979 36 37 38 3. Regulation of lipid metabolism_RXR-dependent regulation of lipid metabolism via PPAR, RAR, and 39 40 VDR (p = 5.1*10 3; Supplemental Table S6). 41
42 on October 1, 2021 by guest. Protected copyright. 43 SNPid Gene Symbol Gene Name OR P-value 44 45 carnitine rs12364396 CPT1A 1.779 0.006433 46 palmitoyltransferase 1A 47 rs1286767;rs216 retinoic acid receptor, 48 RARB 1.647 0.005116 49 4360;rs2033447 beta 50 51 52 3 53 4. BMP (bone morphogenic protein) signaling pathway (p = 5.2*10 ). BMPs are members of the TGF β 54 55 superfamily, with important roles in embryonic development including cardiomyogenesis and apoptosis. 56 57 BMPs transduce their signals via 2 pathways: SMAD dependent (including SMAD9[8]), and SMAD 58 59 60 18
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 51 BMJ Open
1 independent cascade. BMP activity can be regulated by intracellular inhibitory SMADs (like SMAD6) or BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 extracellularly by secreted antagonists (like gremlin). Variants in SMAD9 and SMAD6 were associated 4 5 with an increased incidence of perioperative MI (Supplemental Table S7). 6 7 8 9 SNPid Gene Symbol Gene Name OR p-value 10 11 rs11858577 SMAD6 SMAD, mothers against DPP homolog 6 2.30 0.0002506 12 Mothers against decapentaplegic, drosophila, rs7998663 SMAD9 2.47 0.000167 13 homolog of, 9 14 15 rs884940For SKI peerski oncoprotein review only0.61 0.0006749 16 17 18 19 In summary, functional analysis of genome wide association data using biological ontologies allowed us 20 21 to supplement our single marker analysis results by identifying a number of additional candidates for 22 23 further investigation as biomarkers for perioperative MI. It also highlighted the potential mechanistic 24 25 importance of several signaling pathways that have not previously been implicated in the complex and 26 27 multifactorial pathogenesis of perioperative ischemia reperfusion injury. 28 29 30 31 32 Supplemental Figure S7 summarizes RegulomeDB (regulome.stanford.edu) and HaploReg v2 33 34 (http://www.broadinstitute.org/mammals/haploreg/haploreg.php) database results for rs609408 (intergenic http://bmjopen.bmj.com/ 35 36 variant, located 1.5kb 3’ of SMAD9). 37 38 A. RegulomeDB supporting data for rs609418 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 51
1 B. HaploReg v2 summary for rs609418 and variants with r2≥0.8 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 and detailed view for rs609418 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 20
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 51 BMJ Open
1 B.6. Power calculation BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 We used Genetic Power Calculator (http://pngu.mgh.harvard.edu/~purcell/gpc/) to estimate the effect size 4 5 that our study sample size can detect with 80% power. As an example, we used rs4834703 in the SEC24D 6 7 8 gene, which has 0.1 MAF and odds ratio (OR) of 1.98 from the stage II analysis. In this power calculation, 9 10 we assumed 11% disease prevalence, 0.1 MAF for the marker, and the same study sample sizes (225 11 12 cases and 1830 controls). We varied the minor allele frequency of the disease locus (0.05, 0.1, 0.15) to 13 14 assess the genotypic relative risk (GRR) that our study sample size can detect with 80% power. We used 15 For peer review only 16 the same significant threshold described in the paper, 9 x 10 5, for the stage II analysis. We assumed an 17 18 additive model to define GRR of Aa and AA genotypes, that is, GRR_Aa = x and GRR_Aa = 2x, where x 19 20 21 is equivalent to the OR for the minor allele. If the marker tested was in complete linkage disequilibrium 22 23 (D’ = 1) with the disease locus, we found that our sample size has approximately 80% power to detect a 24 25 GRR = 2.6 for a rare disease locus (MAF = 0.05) and GRR = 1.8 for a more common disease locus (MAF 26 27 = 0.1 and 0.15). The effect sizes become slightly higher when D’ is assumed at 0.8 between disease locus 28 29 and marker (Supplemental Table S8). The OR for rs4834703 in the SEC24D gene in our study was 1.97, 30 31 which is consistent with the estimate here. Although somewhat controversial, current evidence supports 32 33 34 joint 2 stage analysis designs over replication analyses in GWAS, based on increased statistical power of http://bmjopen.bmj.com/
35 5, 6 36 joint analyses, which formed the rationale for our study design. 37 38 Supplemental Table S8. GRR estimates that our dataset can detect a given disease locus MAF at 39 40 approximately 80% power when marker MAF = 0.1, sample size = 225 cases and 1830 controls, and 41 42 disease prevalence = 11%. on October 1, 2021 by guest. Protected copyright. 43 44 45 D' Disease locus MAF Odds Ratio Power 46 1 0.05 2.6 0.8 47 0.1 1.8 0.84 48 49 0.15 1.75 0.81 50 0.8 0.05 3.1 0.8 51 0.1 2.15 0.84 52 0.15 2 0.82 53 54 55 56 57 58 59 60 21
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 51
1 B.7. Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 4 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 5 8 RS2044061 19723503 INTS10 | LPL G 0.3117 2.004 1.499 2.68 2.76E-06 4 RS6839725 21870314 LOC100131836 | KCNIP4 A 0.09312 2.101 1.42 3.11 0.0002053 1 RS6660197 200703541 DDX59 | CAMSAP1L1 A 0.1266 2.368 1.63 3.438 5.99E-06 11 RS1816775 13783230 FAR1 | SPON1 A 0.3728 1.71 1.288 2.269 0.0002075 6 1 RS12756886 200840467 CAMSAP1L1 | GPR25 G 0.1215 2.342 1.612 3.404 8.12E-06 4 RS10516398 21746541 KCNIP4 A 0.09791 2.085 1.414 3.074 0.0002086 3 RS6788848 103028245 LOC100287880 | ALCAM G 0.237 1.995 1.473 2.702 8.15E-06 15 RS4288951 27751567 GABRG3 A 0.08947 2.283 1.476 3.532 0.0002089 7 1 RS12060387 210819848 HHAT A 0.05196 3.218 1.925 5.379 8.27E-06 3 RS9883142 7017919 GRM7 A 0.3052 1.717 1.29 2.285 0.0002102 1 RS12742404 200811765 CAMSAP1L1 G 0.1204 2.338 1.609 3.397 8.44E-06 3 RS12631018 116501764 LOC285194 | IGSF11 G 0.1684 1.913 1.358 2.696 0.0002105 8 6 RS3861449 148300251 SAMD5 | SASH1 A 0.2655 1.985 1.467 2.686 8.77E-06 7 RS17171847 41370672 C7orf10 | INHBA G 0.09312 2.121 1.425 3.157 0.0002106 15 RS894157 90168108 C15orf42 G 0.1012 2.333 1.604 3.394 9.42E-06 15 RS6494826 70049111 LOC145837 | C15orf50 A 0.4318 1.738 1.297 2.33 0.0002164 9 8 RS2235118 2027154 MYOM2 C 0.1319 2.184 1.536 3.107 1.38E-05 10 RS984668 124120559 BTBD16 | PLEKHA1 A 0.3542 1.736 1.296 2.325 0.0002167 1 RS2292096 200826769 CAMSAP1L1 G 0.1279 2.282 1.569 3.318 1.57E-05 17 RS10512598 73038132 ATP5H A 0.05634 2.532 1.547 4.144 0.0002202 10 4 RS4689485 6578487 MAN2B2 A 0.1606 2.054 1.479 2.851 1.71E-05 2 RS7594398 240250456 HDAC4 A 0.0695 2.28 1.47 3.536 0.000232 5 RS7721080 31973116 PDZD2 G 0.04521 3.173 1.868 5.391 1.94E-05 1 RS12032817 83177365 LPHN2 | TTLL7 G 0.3934 1.703 1.283 2.261 0.0002321 11 6 RS9382274 53877181 LRRC1 | C6orf142 A 0.07399 2.601 1.673 4.044 2.20E-05 5 RS17105323 146114766 PPP2R2B G 0.03205 3.205 1.723 5.96 0.0002336 3 RS1038517 59949140 FHIT A 0.0898 2.477 1.622 3.782 2.68E-05 12 RS11171580 55975171 OR6C4 | OR10P1 C 0.2335 1.772 1.306 2.404 0.0002349 12 5 RS331706 124890054 ZNF608 | GRAMD3 A 0.1478 2.128 1.495 3.027 2.71E-05 1 RS821717 232059805 DISC1 C 0.2107 0.4419 0.2859 0.683 0.0002365 4 RS4834703 119691624 SEC24D A 0.09379 2.271 1.544 3.342 3.13E-05 13 RS17424088 23087983 FGF9 | LOC646201 A 0.1055 2.068 1.404 3.046 0.0002365 13 4 RS4302525 6552201 PPP2R2C | MAN2B2 A 0.1457 2.039 1.458 2.853 3.18E-05 2 RS17027173 103057043 IL18RAP A 0.2156 1.814 1.32 2.493 0.0002383 13 RS9547679 37402440 RFXAP A 0.07962 2.63 1.662 4.162 3.63E-05 16 RS16956168 56249774 DKFZP434H168 | GNAO1 A 0.1738 1.887 1.345 2.649 0.0002397 14 1 RS12723891 217143384 ESRRG G 0.1103 2.126 1.479 3.056 4.62E-05 13 RS7332913 79933284 RBM26 G 0.4909 0.5783 0.4317 0.7748 0.0002424 19 RS3745752 48339300 CRX G 0.4011 1.844 1.373 2.476 4.84E-05 13 RS2763810 79979679 RBM26 A 0.4906 0.5783 0.4317 0.7748 0.0002424 15 22 RS138195 46151880 ATXN10 G 0.371 1.819 1.362 2.429 5.04E-05 7 RS2700976 37400570 ELMO1 A 0.2302 1.756 1.3 2.373 0.0002431 6 RS2876669 148216298 SAMD5For | SASH1 A peer0.2328 1.891 1.389 2.5 73 review5.12E-05 14 RS8009218 100057672 CCDC85ConlyG 0.4541 1.733 1.292 2.325 0.0002438 16 4 RS4631029 21796070 KCNIP4 | NCRNA00099 G 0.09885 2.214 1.507 3.253 5.14E-05 2 RS17015041 78416572 SNAR-H | REG3G G 0.1397 1.975 1.373 2.841 0.0002458 20 RS6067635 49698104 MULTIPLE_GENES:3755,100128598 A 0.2601 1.903 1.393 2.6 5.24E-05 22 RS6007696 47968450 LOC100289420 | FAM19A5 G 0.1765 1.878 1.341 2.631 0.0002459 17 16 RS12920403 19756059 IQCK G 0.1235 2.109 1.464 3.036 6.06E-05 9 RS10809097 1057797 DMRT2 | SMARCA2 G 0.1373 1.999 1.38 2.895 0.0002483 6 RS9405142 1632299 GMDS G 0.01484 5.575 2.4 12.95 6.46E-05 15 RS11858577 67066563 SMAD6 A 0.07422 2.3 1.473 3.591 0.0002506 18 9 RS12115781 112672877 MULTIPLE_GENES:445815,114299 C 0.08401 2.359 1.545 3.601 6.97E-05 13 RS2147568 79991226 RBM26 | NDFIP2 G 0.4906 0.5797 0.4329 0.7762 0.0002516 9 RS7042036 119766969 ASTN2 G 0.3168 1.81 1.351 2.425 7.13E-05 10 RS7910400 6310176 PFKFB3 | LOC399715 G 0.4224 0.5523 0.4018 0.7591 0.000254 19 19 RS11668505 48348363 CRX | SULT2A1 G 0.3984 1.803 1.347 2.414 7.43E-05 2 RS3828312 100184719 AFF3 A 0.2045 1.82 1.32 2.508 0.0002564 7 RS10270298 104210910 LHFPL3 A 0.04825 3.011 1.743 5.2 7.76E-05 12 RS10859508 93919441 MRPL42 | SOCS2 A 0.3552 0.5428 0.3911 0.7533 0.0002576 20 14 RS10140464 49210639 MDGA2 | RPS29 G 0.198 1.894 1.38 2.6 7.82E-05 4 RS6822035 119710421 SEC24D A 0.2955 1.734 1.29 2.329 0.0002593 6 RS9352851 81392952 BCKDHB | FAM46A A 0.3728 1.787 1.34 2.384 7.82E-05 9 RS7855908 71517082 PIP5K1B G 0.05466 2.712 1.588 4.631 0.0002598 21 7 RS10953524 106618074 PIK3CG | PRKAR2B G 0.499 0.5472 0.4056 0.7381 7.86E-05 7 RS2723520 17813677 AHR | SNX13 G 0.4521 1.719 1.285 2.299 0.0002609 21 RS928873 43293371 PRDM15 A 0.278 0.4756 0.3287 0.6881 8.05E-05 3 RS967331 61447737 FHIT | PTPRG A 0.1559 1.902 1.347 2.686 0.0002631 22 7 RS2299146 42033195 GLI3 A 0.302 1.763 1.33 2.337 8.10E-05 7 RS11763025 21920165 DNAH11 A 0.3502 1.692 1.275 2.244 0.0002649 8 RS6586870 19709775 INTS10 | LPL A 0.3934 1.8 1.344 2.412 8.11E-05 20 RS2425847 44907927 CDH22 | SLC35C2 G 0.3856 1.706 1.28 2.274 0.0002664 23 1 RS12122147 74838630 TNNI3K G 0.1004 2.181 1.48 3.214 8.14E-05 1 RS10489478 176566350 PAPPA2 A 0.1734 0.4052 0.2493 0.6586 0.0002668 9 RS457417 4855514 RCL1 C 0.1306 2.027 1.426 2.883 8.27E-05 10 RS1915613 66483216 LOC100287799 | ANXA2P3 A 0.248 0.4801 0.3235 0.7124 0.0002689 24 15 RS11638941 39481950 LOC100289563 | C15orf54 G 0.1424 2.008 1.418 2.843 8.47E-05 23 RS7051093 90319063 LOC100288789 | PABPC5 A 0.1842 0.2732 0.1359 0.5491 0.0002701 7 RS2058894 97296791 ACN9 | TAC1 A 0.0368 3.17 1.783 5.636 8.52E-05 6 RS6456672 25347733 LRRC16A C 0.01417 6.505 2.374 17.82 0.0002709 25 17 RS11077582 69911104 LOC124685 | SOX9 A 0.06309 2.456 1.568 3.845 8.62E-05 23 RS5940927 90309551 LOC100288789 | PABPC5 A 0.1831 0.2738 0.1362 0.5504 0.000277 8 RS13254568 11514144 BLK | GATA4 A 0.1732 1.978 1.406 2.783 8.91E-05 23 RS10522055 90307072 LOC100288789 | PABPC5 G 0.1831 0.2738 0.1362 0.5504 0.000277 26 20 RS2425810 44890910 CDH22 | SLC35C2 A 0.3924 1.787 1.335 2.391 9.38E-05 20 RS2425807 44881545 CDH22 | SLC35C2 G 0.4136 1.714 1.281 2.291 0.0002808 10 RS3847355 71725020 COL13A1 | H2AFY2 A 0.442 1.789 1.336 2.396 9.57E-05 8 RS4510897 72277458 EYA1 | MSC G 0.1552 1.898 1.343 2.682 0.0002813 27 5 RS10078649 179877533 GFPT2 | CNOT6 A 0.09818 2.174 1.471 3.212 9.80E-05 8 RS10808467 115320834 CSMD3 | TRPS1 A 0.4899 0.5774 0.4291 0.7768 0.0002854 21 RS4276100 43275896 LOC100288232 | PRDM15 G 0.2652 1.823 1.348 2.467 9.91E-05 22 RS5754826 34476759 LARGE | ISX A 0.2893 0.5043 0.3483 0.73 0.0002861 28 2 RS11686987 107128885 RGPD3 | LOC644604 A 0.03038 3.33 1.817 6.105 9.97E-05 9 RS752357 36744752 MELK | PAX5 G 0.3478 0.5406 0.3877 0.7538 0.0002875 16 RS8055711 7040933 A2BP1 A 0.2881 1.825 1.348 2.47 1.00E-04 1 RS10489477 176562608 PAPPA2 A 0.169 0.3777 0.2231 0.6396 0.0002909 29 2 RS4853374 78442014 SNAR-H | REG3G A 0.1353 2.076 1.436 3.002 0.0001033 9 RS11792612 17442000 CNTLN A 0.1798 1.812 1.314 2.5 0.0002932 19 RS7260368 57135195 ZNF71 G 0.4855 1.756 1.321 2.335 0.0001076 10 RS12780002 6308484 PFKFB3 | LOC399715 A 0.3342 0.5359 0.3821 0.7514 0.0002986 30 6 RS9361640 81392005 BCKDHB | FAM46A C 0.3013 1.793 1.333 2.412 0.0001118 8 RS6997580 115321807 CSMD3 | TRPS1 C 0.4882 0.5787 0.4301 0.7787 0.000305 10 RS2895524 12595147 CAMK1D A 0.3465 0.5135 0.3662 0.7201 0.000112 3 RS1284745 102324326 ZPLD1 | LOC100287880 G 0.3313 1.714 1.279 2.297 0.0003073 31 14 RS17112396 42397496 LRFN5 | FSCB A 0.01316 6.11 2.436 15.32 0.0001143 1 RS10801786 90595332 ZNF326 | BARHL2 C 0.1781 1.816 1.313 2.511 0.0003103 8 RS7839240 39746386 ADAM2 | IDO1 G 0.3998 1.763 1.322 2.353 0.0001152 13 RS609418 37417427 RFXAP | SMAD9 C 0.2551 1.731 1.284 2.333 0.0003138 32 4 RS2880099 156432368 LOC100287564 | GUCY1A3 A 0.1191 2.059 1.426 2.974 0.0001162 6 RS4708617 168389858 HGC6.3 | KIF25 A 0.1134 0.2382 0.1091 0.5198 0.0003142 2 RS4951964 31300544 GALNT14 A 0.02767 3.535 1.86 6.719 0.0001165 4 RS3108140 107334044 LOC100288276 A 0.3107 1.7 1.273 2.269 0.0003187 33 14 RS1242558 83235421 SEL1L | FLRT2 A 0.2912 1.785 1.329 2.398 0.0001186 2 RS1370631 78410008 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233
15 RS12913680 90098150 LOC283761 | C15orf42 A 0.1093 2.144 1.452 3.164 0.0001236 2 RS2901929 78428973 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233 http://bmjopen.bmj.com/ 34 2 RS2894595 227237862 KIAA1486 | IRS1 C 0.2895 1.77 1.322 2.37 0.0001245 2 RS2164850 78423757 SNAR-H | REG3G A 0.141 1.949 1.355 2.803 0.0003233 4 RS1398835 21849857 NCRNA00099 A 0.09987 2.12 1.444 3.113 0.0001262 16 RS4786590 5125365 ALG1 G 0.0135 6.207 2.291 16.81 0.0003296 35 9 RS7035935 112706482 PALM2 C 0.06748 2.47 1.555 3.923 0.0001278 14 RS4902165 63115447 FLJ43390 | KCNH5 G 0.1387 1.921 1.345 2.744 0.0003297 2 RS11686479 31309930 GALNT14 A 0.028 3.505 1.845 6.66 0.0001284 6 RS7753022 86176625 NT5E A 0.1397 1.975 1.362 2.864 0.0003344 36 7 RS6461168 2170759 MAD1L1 G 0.2756 1.766 1.318 2.365 0.0001368 21 RS2839401 43302079 PRDM15 | C2CD2 A 0.2402 0.4871 0.3288 0.7217 0.0003358 13 RS6563501 37408545 RFXAP | SMAD9 G 0.2257 1.827 1.34 2.491 0.0001385 9 RS11144511 78218231 OSTF1 | PCSK5 A 0.4501 1.71 1.275 2.293 0.0003408 37 12 RS11111722 104104227 STAB2 A 0.05162 2.752 1.634 4.633 0.0001398 11 RS12799462 13977738 FAR1 | SPON1 C 0.2034 0.4419 0.2825 0.6911 0.0003456 2 RS11685519 107122121 RGPD3 | LOC644604 A 0.03138 3.156 1.746 5.702 0.0001408 6 RS12192975 20431741 E2F3 A 0.05634 2.357 1.474 3.769 0.0003457 38 7 RS4346898 21951033 CDCA7L A 0.391 1.715 1.299 2.264 0.0001422 23 RS180484 150318158 HMGB3 | LOC100128688 G 0.3481 1.961 1.356 2.837 0.0003485 21 RS870531 43276783 LOC100288232 | PRDM15 A 0.3292 0.5185 0.3697 0.7273 0.0001424 10 RS161016 110039615 LOC100128304 | LOC645318 G 0.3381 0.5504 0.3967 0.7635 0.0003497 39 5 RS716866 168645723 SLIT3 G 0.4356 1.741 1.308 2.317 0.0001427 16 RS7188523 55313079 LOC100132339 | IRX6 A 0.04757 2.643 1.551 4.504 0.0003502 5 RS6595621 124989579 ZNF608 | GRAMD3 A 0.2375 1.858 1.35 2.556 0.0001429 12 RS2303970 74471340 TRHDE | LOC552889 A 0.4673 0.5823 0.4328 0.7833 0.0003512 40 1 RS1183044 209255814 PLXNA2 | LOC642587 A 0.4922 1.744 1.309 2.323 0.000146 5 RS297853 168628678 SLIT3 A 0.3367 1.683 1.265 2.239 0.0003513 7 RS4140805 7727101 RPA3 C 0.4214 1.747 1.309 2.33 0.0001501 4 RS1908125 134392080 PCDH10 | PABPC4L G 0.2068 1.757 1.29 2.394 0.0003531 41 5 RS10447192 8896731 LOC100128382 | SEMA5A A 0.4069 0.5487 0.4022 0.7486 0.0001528 7 RS2237429 42031224 GLI3 A 0.1883 1.791 1.301 2.466 0.0003571 2 RS858940 50945699 NRXN1 A 0.3836 1.743 1.307 2.324 0.0001535 8 RS6989246 2015119 MYOM2 A 0.09514 2.071 1.389 3.088 0.0003573 on October 1, 2021 by guest. Protected copyright. 42 7 RS10263415 7724805 LOC729852 | RPA3 A 0.4244 1.755 1.312 2.349 0.000154 8 RS7820007 142731821 FLJ43860 | NCRNA00051 A 0.3921 1.695 1.269 2.264 0.0003578 10 RS17452064 11242471 CUGBP2 C 0.1088 2.059 1.416 2.995 0.0001571 6 RS6932311 25346376 LRRC16A A 0.01383 6.219 2.279 16.97 0.0003597 43 15 RS8041127 90135196 C15orf42 A 0.1043 2.118 1.435 3.128 0.0001596 22 RS138204 46161875 ATXN10 A 0.3647 1.686 1.266 2.247 0.0003607 6 RS6456676 25350468 LRRC16A C 0.01181 6.973 2.543 19.12 0.0001614 12 RS2365432 74439038 TRHDE | LOC552889 A 0.4136 0.5733 0.4223 0.7783 0.0003619 44 7 RS3778973 2151178 MAD1L1 A 0.208 1.809 1.329 2.462 0.0001653 15 RS4777184 70007323 LOC145837 | C15orf50 A 0.365 1.704 1.271 2.284 0.0003629 13 RS7998663 37435150 SMAD9 G 0.07456 2.47 1.542 3.954 0.000167 7 RS10233681 71941439 CALN1 | TYW1B A 0.2554 0.4969 0.3382 0.73 0.000366 45 8 RS10086641 39754780 ADAM2 | IDO1 A 0.4035 1.738 1.303 2.318 0.0001693 4 RS6828577 119704501 SEC24D A 0.2939 1.713 1.274 2.302 0.0003672 4 RS1878943 9766888 LOC100128096 | DRD5 A 0.1926 1.848 1.342 2.546 0.0001704 7 RS17410662 96231614 FLJ42280 | SHFM1 A 0.1427 1.877 1.327 2.655 0.0003686 46 6 RS220702 46867717 GPR116 G 0.09885 2.161 1.446 3.231 0.0001718 6 RS744143 20418718 E2F3 A 0.07596 2.156 1.412 3.29 0.0003701 11 RS554341 128685761 FLI1 | KCNJ1 G 0.1334 1.946 1.374 2.754 0.0001746 20 RS6033188 11585276 JAG1 | LOC728573 G 0.1353 0.3323 0.1812 0.6095 0.0003707 47 20 RS7508996 40396739 CHD6 | PTPRT G 0.08603 2.294 1.487 3.539 0.000175 5 RS4298224 8894289 LOC100128382 | SEMA5A A 0.3698 0.5636 0.411 0.7729 0.0003729 10 RS2184035 19778072 C10orf112 A 0.04453 2.7 1.606 4.538 0.0001773 2 RS10179116 100174645 AFF3 A 0.3522 1.714 1.273 2.306 0.0003763 48 7 RS7802845 111723488 DOCK4 G 0.07794 2.318 1.493 3.598 0.0001797 12 RS10431428 93898858 MRPL42 | SOCS2 G 0.2831 1.769 1.292 2.422 0.0003775 12 RS1316899 55966101 OR6C4 | OR10P1 A 0.1619 1.844 1.339 2.541 0.0001813 15 RS2289486 90161355 C15orf42 G 0.09177 2.068 1.385 3.086 0.0003776 49 1 RS10454444 176649181 PAPPA2 G 0.1754 0.3867 0.2352 0.6359 0.0001814 7 RS1404508 96228434 FLJ42280 | SHFM1 A 0.1518 1.855 1.32 2.608 0.0003779 7 RS17482047 7755484 RPA3 G 0.08835 2.177 1.448 3.272 0.0001832 4 RS6854328 119683492 SEC24D A 0.2939 1.711 1.272 2.3 0.0003788 50 1 RS10913237 176630416 PAPPA2 A 0.1748 0.3866 0.2349 0.6362 0.0001846 2 RS13030183 79909989 CTNNA2 A 0.4467 1.681 1.262 2.239 0.0003823 15 RS4774964 57854873 CGNL1 | GCOM1 G 0.1859 0.4038 0.2508 0.65 0.0001892 9 RS10969620 30323457 LOC100289571 | LOC100288436 A 0.0948 1.983 1.358 2.894 0.0003898 51 12 RS3871519 93915922 MRPL42 | SOCS2 A 0.3526 0.5349 0.3851 0.7429 0.0001894 6 RS6568370 105845531 PREP C 0.1087 2.048 1.378 3.045 0.0003918 2 RS1595422 78434830 SNAR-H | REG3G A 0.1326 2.03 1.4 2.943 0.0001897 2 RS12469727 227301483 KIAA1486 | IRS1 A 0.1653 1.819 1.307 2.532 0.0003923 52 4 RS298988 119648399 SEC24D A 0.226 1.793 1.32 2.437 0.0001899 2 RS13422448 49743324 FSHR | NRXN1 C 0.3188 0.5325 0.3758 0.7545 0.0003934 4 RS1491152 130547852 C4orf33 | LOC100132483 G 0.3347 1.74 1.3 2.329 0.0001949 23 RS5941802 90288180 LOC100288789 | PABPC5 C 0.179 0.2835 0.1412 0.5693 0.000394 53 20 RS6022863 52556423 SUMO1P1 | BCAS1 A 0.4028 0.5521 0.4038 0.7548 0.0001965 16 RS899236 56352905 GNAO1 A 0.2095 1.789 1.297 2.469 0.0003944 8 RS4875192 3187176 CSMD1 A 0.03914 2.825 1.634 4.883 0.0001996 17 RS767234 69886832 LOC124685 | SOX9 A 0.2186 1.766 1.289 2.42 0.0003964 54 15 RS17738087 26905021 GABRB3 G 0.0975 2.117 1.425 3.145 0.0002022 20 RS3091869 45425918 LOC100127904 | EYA2 A 0.4325 1.665 1.256 2.207 0.0003971 3 RS17061897 60020954 FHIT A 0.0587 2.476 1.534 3.994 0.0002039 16 RS17720036 27139728 C16orf82 | G 0.1309 1.929 1.341 2.775 0.0003997 55 1 RS6577559 9238095 GPR157 | LOC727721 A 0.1231 1.985 1.382 2.85 0.000205 10 RS6593445 45359483 LOC100289168 | TMEM72 G 0.1501 1.854 1.317 2.609 0.0004009 56 57 58 59 60 22
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 51 BMJ Open
1 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 4 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 5 2 RS1356799 140567666 LOC647012 | LRP1B C 0.1977 0.439 0.2783 0.6926 0.0004021 2 RS893745 189088249 LOC729141 G 0.4933 0.5933 0.4406 0.799 0.0005872 3 RS6770667 71873579 PROK2 | RYBP A 0.2903 1.686 1.262 2.251 0.0004067 6 RS6911476 26612363 ABT1 | ZNF322A A 0.02328 4.264 1.864 9.752 0.0005914 6 2 RS6708183 78384746 SNAR-H | REG3G A 0.3252 1.706 1.269 2.295 0.0004101 4 RS6858670 137258380 LOC100289626 | LOC646316 G 0.4656 0.598 0.4459 0.8019 0.0005918 4 RS11730104 4682267 STX18 | LOC100289434 C 0.4136 1.665 1.255 2.209 0.0004125 4 RS7688325 137260052 LOC100289626 | LOC646316 A 0.4656 0.598 0.4459 0.8019 0.0005918 7 8 RS1497194 115351857 CSMD3 | TRPS1 G 0.4899 0.5859 0.4355 0.7883 0.0004131 8 RS290623 115395451 CSMD3 | TRPS1 A 0.473 0.5957 0.4433 0.8006 0.0005939 20 RS6032661 44733811 NCOA5 | CD40 G 0.2208 0.4692 0.3083 0.7142 0.0004147 16 RS9889027 81341363 BCMO1 | GAN G 0.2362 1.703 1.257 2.308 0.0005983 8 20 RS6064024 52546889 SUMO1P1 | BCAS1 G 0.4265 1.682 1.26 2.244 0.0004162 16 RS8055894 21190092 TMEM159 A 0.1414 1.894 1.315 2.728 0.0006004 11 RS7926320 13782568 FAR1 | SPON1 A 0.388 1.664 1.254 2.207 0.0004164 7 RS1859107 21859436 DNAH11 C 0.1333 0.367 0.207 0.6506 0.0006012 9 15 RS8043317 27751786 GABRG3 A 0.05466 2.493 1.501 4.141 0.000418 9 RS2417753 109405667 LOC644620 G 0.1549 1.845 1.3 2.618 0.0006025 17 RS1126642 42989063 GFAP A 0.03552 2.741 1.565 4.8 0.0004211 19 RS296366 48374538 SULT2A1 A 0.238 1.728 1.264 2.361 0.0006055 10 10 RS317499 109995284 LOC100128304 | LOC645318 A 0.3357 0.5558 0.4009 0.7706 0.0004257 4 RS10516407 21883570 LOC100131836 | KCNIP4 A 0.1891 1.791 1.284 2.499 0.0006057 15 RS4777990 93018587 C15orf32 G 0.2817 1.715 1.27 2.316 0.0004304 10 RS3818540 97165811 SORBS1 A 0.1043 2.08 1.369 3.162 0.0006058 11 2 RS10495786 33559429 LTBP1 G 0.1073 1.974 1.352 2.884 0.0004312 10 RS10733970 58773025 ZWINT | IPMK A 0.02161 5.501 2.075 14.58 0.0006088 2 RS10200850 240234901 HDAC4 A 0.0652 2.22 1.424 3.462 0.0004344 1 RS6673391 30731688 LOC100288450 | MATN1 G 0.108 2.04 1.357 3.068 0.0006099 12 3 RS9868354 108840227 MORC1 | LOC100288721 G 0.02901 3.265 1.689 6.313 0.0004352 5 RS17192980 6977768 POLS | LOC442132 G 0.09959 1.951 1.331 2.859 0.0006113 2 RS1922351 35300334 MYADML | LOC100288911 G 0.3924 1.681 1.259 2.246 0.0004369 10 RS4750200 6308646 PFKFB3 | LOC399715 A 0.3404 0.556 0.3974 0.7779 0.0006118 13 16 RS2356417 71356135 FTSJD1 | CALB2 G 0.1265 1.927 1.337 2.777 0.0004381 8 RS987341 18009367 ASAH1 | NAT1 A 0.05908 2.282 1.423 3.658 0.000613 15 RS2028465 71659358 THSD4 C 0.2156 1.771 1.288 2.437 0.000441 16 RS7203768 73301302 LOC100288121 | LOC401859 G 0.3962 0.5831 0.4282 0.7939 0.0006145 14 2 RS888281 56507299 CCDC85A G 0.3252 1.662 1.252 2.208 0.0004457 18 RS9953631 39047265 LOC647946 | KC6 G 0.3239 0.5518 0.3926 0.7755 0.0006176 23 RS12559251 90342188 LOC100288789 | PABPC5 C 0.175 0.2844 0.1409 0.5738 0.0004462 12 RS10770407 19087150 CAPZA3 | PLEKHA5 G 0.1381 1.939 1.327 2.833 0.0006179 15 16 RS1424106 79216793 WWOX ForA 0.4686peer 0.5839 0.4324 0.7884 0.0004463review8 RS7819086 115272300 CSMD3 only| TRPS1 A 0.427 1.645 1.237 2.188 0.0006201 2 RS17553946 155347161 GALNT13 | KCNJ3 A 0.2236 0.4806 0.3192 0.7236 0.000449 1 RS381115 153099345 SPRR2F | SPRR2C A 0.1171 1.965 1.335 2.892 0.0006205 16 10 RS12220134 97129606 SORBS1 A 0.06174 2.421 1.477 3.966 0.0004499 3 RS4927870 195820576 TFRC | FLJ25996 A 0.1191 1.975 1.338 2.917 0.0006207 2 RS16859880 220267737 DNPEP | LOC100288941 A 0.02126 4.044 1.853 8.828 0.0004507 12 RS4761520 93910308 MRPL42 | SOCS2 G 0.3067 0.5467 0.3869 0.7726 0.0006213 17 2 RS1922353 35302429 MYADML | LOC100288911 C 0.418 1.689 1.26 2.264 0.0004518 19 RS3786656 14506664 CD97 A 0.3364 1.645 1.237 2.188 0.0006225 21 RS2830247 27848998 CYYR1 G 0.07625 2.105 1.388 3.193 0.0004582 10 RS2171302 3815292 LOC100287172 A 0.1721 1.826 1.293 2.578 0.000624 18 9 RS7036825 16125443 C9orf93 | BNC2 C 0.02193 3.692 1.778 7.666 0.0004596 10 RS624964 11231831 CUGBP2 G 0.3161 1.687 1.25 2.276 0.0006241 20 RS2868714 44886189 CDH22 | SLC35C2 G 0.4879 1.676 1.255 2.237 0.0004617 2 RS7594456 212403434 ERBB4 A 0.3185 1.659 1.241 2.218 0.0006277 19 13 RS9516888 97931709 MBNL2 A 0.4308 1.645 1.245 2.174 0.0004621 8 RS3808606 39769375 ADAM2 | IDO1 A 0.4727 1.659 1.241 2.217 0.0006282 10 RS3932533 3925728 LOC100130652 | LOC727894 A 0.277 0.5138 0.3538 0.7461 0.000468 17 RS7215344 2430126 METT10D | PAFAH1B1 G 0.2505 0.5055 0.3418 0.7476 0.0006328 20 9 RS2503351 19182893 ADFP | LOC100288002 G 0.1694 1.826 1.303 2.558 0.0004689 17 RS7222013 66267691 SLC16A6 T 0.2498 1.737 1.265 2.384 0.0006334 19 RS10469470 17187518 MYO9B C 0.1276 0.3353 0.1817 0.6185 0.0004691 10 RS317490 110051491 LOC100128304 | LOC645318 G 0.3357 0.5664 0.4088 0.7848 0.0006345 21 13 RS12869943 38356247 TRPC4 G 0.2549 0.5119 0.3517 0.745 0.0004698 6 RS9463046 44802718 SUPT3H G 0.03981 2.735 1.536 4.872 0.0006349 2 RS328632 9247782 MBOAT2 | ASAP2 A 0.1694 0.3992 0.2386 0.6679 0.0004703 8 RS290592 115370512 CSMD3 | TRPS1 G 0.4835 0.5978 0.445 0.8031 0.0006351 22 7 RS2158499 88558769 C7orf62 | ZNF804B A 0.4163 0.5763 0.4231 0.7849 0.000472 11 RS1528663 13967222 FAR1 | SPON1 G 0.1916 0.4625 0.2971 0.7199 0.0006367 10 RS12766986 3335061 PITRM1 | LOC100287172 G 0.1842 1.764 1.283 2.424 0.0004726 11 RS1507526 13940539 FAR1 | SPON1 A 0.363 1.658 1.24 2.216 0.0006371 23 13 RS6491347 97930890 MBNL2 A 0.4372 1.641 1.243 2.167 0.0004732 8 RS4876325 115239083 CSMD3 | TRPS1 G 0.448 1.641 1.235 2.18 0.0006374 6 RS6910470 149552517 UST | MAP3K7IP2 A 0.05815 2.367 1.46 3.836 0.0004738 6 RS1333224 137801943 IFNGR1 | OLIG3 A 0.1134 1.968 1.334 2.902 0.0006387 24 7 RS2160138 7755797 RPA3 G 0.4565 1.676 1.255 2.239 0.0004741 15 RS8030303 61772892 RORA | VPS13C A 0.02328 3.421 1.689 6.93 0.0006395 12 RS1357241 74509618 TRHDE | LOC552889 C 0.4821 0.5933 0.4426 0.7952 0.0004769 15 RS4906904 27034085 GABRB3 | GABRA5 A 0.07119 2.198 1.399 3.455 0.0006402 25 22 RS929020 47971041 LOC100289420 | FAM19A5 A 0.17 1.836 1.306 2.583 0.0004774 1 RS563189 83284533 LPHN2 | TTLL7 A 0.4204 1.634 1.233 2.167 0.0006456 20 RS2425835 44901836 CDH22 | SLC35C2 G 0.4337 1.673 1.253 2.234 0.0004801 3 RS6437813 108734088 MORC1 A 0.03036 3.124 1.623 6.011 0.0006474 26 9 RS11143833 71433741 PIP5K1B A 0.01757 3.784 1.793 7.987 0.0004811 10 RS12570530 12547825 CAMK1D G 0.2901 0.5473 0.3871 0.7739 0.0006483 22 RS7284093 51121521 SHANK3 A 0.1492 1.876 1.318 2.671 0.000482 10 RS11591904 45346953 LOC100289168 | TMEM72 G 0.2783 1.678 1.246 2.26 0.0006578 27 18 RS2564500 57230260 CCBE1 G 0.3822 1.668 1.252 2.224 0.0004825 20 RS6127015 52540606 SUMO1P1 | BCAS1 A 0.4585 0.5941 0.4403 0.8017 0.0006598 3 RS212020 59988905 FHIT A 0.09615 2.096 1.383 3.176 0.0004827 19 RS182420 48372195 CRX | SULT2A1 G 0.2407 1.722 1.259 2.354 0.0006606 28 6 RS3869129 31410649 MICA | HCP5 A 0.2142 1.757 1.28 2.411 0.000485 10 RS6585876 86450858 LOC100131699 | GRID1 A 0.335 1.649 1.236 2.199 0.0006643 15 RS1520015 37429504 MEIS2 | TMCO5A A 0.2085 0.4596 0.2969 0.7114 0.0004863 12 RS1800159 57593894 LRP1 A 0.3188 1.646 1.235 2.193 0.0006646 29 6 RS17800315 150765510 IYD | PLEKHG1 A 0.2419 1.709 1.265 2.309 0.0004875 6 RS3101826 160584853 LOC100289162 | SLC22A2 A 0.3333 1.634 1.232 2.168 0.0006649 6 RS3899823 31410597 MICA | HCP5 A 0.2146 1.757 1.28 2.412 0.0004887 20 RS3092379 45432808 LOC100127904 | EYA2 G 0.4342 0.5947 0.4408 0.8023 0.0006705 30 6 RS11757075 150766316 IYD | PLEKHG1 A 0.2421 1.708 1.264 2.309 0.0004914 11 RS161138 110892082 LOC120364 | C11orf53 A 0.3991 0.5867 0.4314 0.7978 0.0006739 12 RS2607916 716509 NINJ2 | LOC100049716 G 0.2301 1.738 1.274 2.372 0.0004928 1 RS884940 2223866 SKI A 0.4298 1.628 1.229 2.155 0.0006749 31 15 RS8030720 90143567 C15orf42 A 0.1005 2.016 1.359 2.99 0.0004954 2 RS13392442 221100327 SLC4A3 | EPHA4 A 0.03374 2.84 1.556 5.183 0.0006756 12 RS4489787 48811100 ZNF641 | ANP32D G 0.1162 0.3285 0.1756 0.6147 0.0004976 7 RS10254402 104192398 LHFPL3 G 0.06849 2.295 1.422 3.706 0.0006758 32 14 RS2369319 96355334 TCL1A | C14orf132 A 0.1775 1.788 1.289 2.48 0.0004985 8 RS3802328 19297051 CSGALNACT1 G 0.3313 0.5617 0.4028 0.7833 0.0006762 20 RS1883838 44761896 CD40 | CDH22 A 0.4126 0.5742 0.4201 0.7848 0.0005009 5 RS256314 34617222 LOC729915 | LOC401180 A 0.03306 3.221 1.64 6.323 0.0006787 33 14 RS12897979 100056208 CCDC85C A 0.368 1.676 1.253 2.243 0.0005036 22 RS5754818 34464710 LARGE | ISX A 0.4217 0.5851 0.4295 0.7971 0.000679 15 RS6576617 27035375 GABRB3 | GABRA5 G 0.1053 2.004 1.354 2.965 0.0005056 2 RS3771166 102986222 IL18R1 A 0.3782 1.654 1.237 2.211 0.0006793 http://bmjopen.bmj.com/ 34 18 RS4797173 466351 COLEC12 A 0.2517 1.707 1.263 2.308 0.0005061 9 RS10816456 98451928 PTCH1 | LOC100287056 A 0.225 0.5024 0.3378 0.7473 0.0006798 12 RS1032725 32641200 BICD1 | FGD4 G 0.3109 1.676 1.253 2.242 0.0005065 17 RS1858990 59735821 NACA2 | BRIP1 A 0.1707 1.783 1.277 2.488 0.0006802 35 3 RS2194938 139731000 CLSTN2 A 0.1955 1.764 1.281 2.43 0.0005077 7 RS10272276 41411165 C7orf10 | INHBA G 0.198 1.75 1.267 2.417 0.000681 12 RS1694747 17617111 LOC390298 | RERGL A 0.08632 2.06 1.37 3.095 0.0005083 14 RS8017842 100055722 CCDC85C G 0.4049 1.657 1.238 2.219 0.0006851 36 3 RS12497400 59986922 FHIT G 0.1193 0.3111 0.161 0.6012 0.0005136 17 RS9910870 1471611 PITPNA | LOC100288617 A 0.2534 0.5119 0.3478 0.7535 0.0006854 18 RS4797604 1213560 LOC100130247 | C18orf2 G 0.1215 2.024 1.359 3.012 0.0005148 9 RS7873155 16118758 C9orf93 | BNC2 C 0.02868 3.443 1.686 7.033 0.0006913 37 3 RS575653 187596668 BCL6 | LOC339929 A 0.2831 1.69 1.257 2.272 0.0005167 6 RS16895923 42633368 UBR2 G 0.1616 1.794 1.28 2.515 0.000692 17 RS7502216 36612948 SOCS7 | ARHGAP23 A 0.3826 0.5652 0.4095 0.7801 0.0005191 4 RS716051 113964288 ANK2 G 0.1614 1.851 1.297 2.642 0.0006941 38 2 RS11888120 29322384 C2orf71 | CLIP4 A 0.2456 1.707 1.262 2.309 0.0005198 18 RS1019868 66753605 CCDC102B | DOK6 C 0.01284 19.66 3.516 109.9 0.0006952 2 RS2390165 166059919 SCN3A A 0.448 1.673 1.251 2.238 0.0005199 2 RS10204137 102968212 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 39 5 RS10041897 169893751 KCNMB1 | KCNIP1 A 0.225 0.4793 0.3164 0.7261 0.0005201 2 RS10192157 102968356 IL1RL1 A 0.3802 1.652 1.236 2.209 0.0006956 20 RS3092167 44922717 CDH22 | SLC35C2 A 0.4578 1.646 1.242 2.182 0.0005235 2 RS10206753 102968362 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 40 18 RS3748415 10471732 LOC100288932 / APCDD1 A 0.1144 0.2917 0.1453 0.5853 0.0005263 9 RS10901213 133247453 HMCN2 C 0.113 1.931 1.32 2.825 0.0006968 5 RS6891322 169897753 KCNMB1 | KCNIP1 A 0.1781 0.4204 0.2575 0.6862 0.0005278 16 RS9924119 71343397 FTSJD1 | CALB2 G 0.1184 1.915 1.315 2.788 0.0006983 41 7 RS17170988 37361023 ELMO1 C 0.2196 0.4803 0.3173 0.7272 0.0005287 15 RS4632084 70049424 LOC145837 | C15orf50 A 0.281 0.535 0.3726 0.7681 0.0006989
6 RS2223451 150767547 IYD | PLEKHG1 A 0.2422 1.701 1.259 2.296 0.0005304 6 RS9497912 148296513 SAMD5 | SASH1 A 0.1269 1.901 1.311 2.756 0.0007016 on October 1, 2021 by guest. Protected copyright. 42 20 RS2425856 44911954 CDH22 | SLC35C2 A 0.4416 1.641 1.24 2.171 0.0005317 2 RS831010 170139346 LRP2 G 0.165 1.792 1.279 2.511 0.0007023 4 RS1438217 118039852 TRAM1L1 | LOC100288955 A 0.4335 1.652 1.243 2.194 0.0005318 12 RS7314861 29238335 CCDC91 | FAR2 A 0.1987 1.777 1.274 2.479 0.0007045 43 12 RS10771958 32638162 BICD1 | FGD4 C 0.3107 1.667 1.248 2.226 0.0005321 16 RS4287581 7032053 A2BP1 G 0.2594 1.703 1.252 2.318 0.0007073 7 RS2058893 97283118 ACN9 | TAC1 A 0.113 1.882 1.316 2.691 0.0005331 10 RS17571397 12572613 CAMK1D A 0.2601 0.5233 0.3595 0.7616 0.0007182 44 3 RS7626013 59961159 FHIT G 0.1076 2.046 1.364 3.069 0.0005428 10 RS4021187 126756083 CTBP2 A 0.4825 0.6061 0.4535 0.8102 0.0007191 9 RS1972384 110210248 RAD23B | KLF4 A 0.1113 1.977 1.343 2.91 0.0005493 2 RS2048765 240208154 HDAC4 G 0.0641 2.183 1.388 3.433 0.0007214 45 15 RS28408948 81998391 TMC3 | MEX3B G 0.0105 7.914 2.448 25.58 0.0005494 2 RS4676312 107658016 LOC649489 | LOC729121 C 0.05162 2.39 1.442 3.961 0.0007214 6 RS476235 160622637 LOC100289162 | SLC22A2 A 0.3428 1.65 1.242 2.193 0.0005521 8 RS925760 115227867 CSMD3 | TRPS1 A 0.4443 1.629 1.227 2.161 0.0007231 46 6 RS1885487 7668923 SNRNP48 | BMP6 G 0.3222 1.71 1.261 2.318 0.0005527 15 RS6494825 70048984 LOC145837 | C15orf50 A 0.4238 0.5973 0.443 0.8053 0.0007234 1 RS12061205 176554721 PAPPA2 A 0.1515 0.3932 0.2315 0.6681 0.0005568 17 RS1971043 20043515 CYTSB G 0.1838 0.4333 0.2668 0.7037 0.0007244 47 1 RS10494201 118252257 FAM46C | LOC100131261 A 0.3151 1.658 1.244 2.209 0.0005574 7 RS11971322 41386423 C7orf10 | INHBA A 0.1866 1.76 1.268 2.442 0.0007256 12 RS1992136 109777361 FOXN4 | MYO1H G 0.1103 0.3186 0.1664 0.6102 0.0005604 12 RS7305820 74550782 TRHDE | LOC552889 G 0.4953 1.635 1.229 2.175 0.0007274 48 7 RS10486908 82860532 PCLO | SEMA3E A 0.05803 2.398 1.459 3.942 0.0005634 10 RS11188295 97125874 SORBS1 A 0.08907 2.149 1.379 3.35 0.0007291 2 RS6744657 155012888 GALNT13 A 0.237 0.4909 0.3276 0.7357 0.0005662 3 RS3772460 59943532 FHIT G 0.1086 2.005 1.339 3.003 0.0007339 49 12 RS10861433 105877556 C12orf75 | NUAK1 A 0.226 0.4811 0.3173 0.7294 0.000568 9 RS295263 4840063 RCL1 G 0.1208 1.869 1.3 2.686 0.0007348 3 RS9860107 60040105 FHIT G 0.07996 2.157 1.393 3.34 0.0005682 18 RS11080417 10423834 VAPA | APCDD1 G 0.2679 1.698 1.249 2.309 0.000736 50 9 RS11103417 137494787 RXRA | COL5A1 G 0.2436 0.5026 0.3398 0.7432 0.0005683 17 RS7220740 59740793 NACA2 | BRIP1 A 0.3133 1.641 1.231 2.188 0.0007379 51 4 RS2196716 118055429 TRAM1L1 | LOC100288955 C 0.4764 0.5955 0.4434 0.7997 0.0005697 2 RS6431565 238836683 RAMP1 | UBE2F A 0.3316 1.67 1.24 2.249 0.0007438 15 RS1435835 26412976 LOC100128714 | GABRB3 G 0.1292 1.931 1.328 2.808 0.0005763 12 RS12823898 93860565 UBE2N | MRPL42 A 0.114 1.978 1.331 2.939 0.0007443 52 21 RS4919983 43301781 PRDM15 | C2CD2 A 0.3155 0.5512 0.3926 0.7737 0.0005765 4 RS1395114 113953581 ANK2 A 0.191 1.797 1.278 2.526 0.0007508 1 RS12067235 205281255 NUAK2 A 0.2213 0.4798 0.3158 0.7291 0.0005813 11 RS627354 74028944 P4HA3 | PGM2L1 A 0.4325 1.621 1.224 2.146 0.0007512 53 8 RS6985983 115289224 CSMD3 | TRPS1 A 0.4265 1.649 1.24 2.193 0.0005824 1 RS10462021 7897133 PER3 G 0.1842 1.744 1.262 2.411 0.0007523 8 RS1552237 87045760 REXO1L2P | PSKH2 A 0.2854 1.672 1.247 2.241 0.0005854 1 RS12125785 209254017 PLXNA2 | LOC642587 A 0.4015 0.595 0.4398 0.8049 0.0007573 54 9 RS10739231 98453680 PTCH1 | LOC100287056 A 0.4281 0.5842 0.43 0.7936 0.0005855 13 RS9531028 80767832 LOC729479 | SPRY2 A 0.1899 0.4483 0.2811 0.7151 0.0007574 12 RS4767252 115187724 TBX3 | MED13L A 0.33 1.67 1.247 2.237 0.0005858 19 RS11083726 29388987 LOC148189 | LOC148145 A 0.4565 1.645 1.231 2.198 0.0007601 55 12 RS995603 29235933 CCDC91 | FAR2 C 0.1993 1.785 1.283 2.484 0.0005862 7 RS4730427 78285960 MAGI2 A 0.1309 0.3623 0.2006 0.6542 0.0007604 7 RS206196 20361873 MACC1 | ITGB8 A 0.4213 0.5846 0.4305 0.7939 0.0005865 1 RS7527871 2281726 MORN1 | LOC100129534 C 0.1715 1.769 1.269 2.465 0.0007626 56 57 58 59 60 23
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 51
1 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 4 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 5 12 RS11046362 22460338 ST8SIA1 A 0.1046 2.003 1.336 3.001 0.000764 5 RS31036 146615260 STK32A C 0.3286 1.644 1.227 2.203 0.0008802 7 RS2204779 118438184 ANKRD7 | LOC100287273 G 0.01619 4.926 1.946 12.47 0.0007653 7 RS974060 97271508 ACN9 | TAC1 G 0.2939 1.643 1.226 2.201 0.0008811 6 8 RS1606891 115206234 CSMD3 | TRPS1 A 0.4483 1.627 1.225 2.16 0.0007696 2 RS10183733 135071570 MGAT5 A 0.1772 1.773 1.265 2.484 0.0008825 12 RS12368128 125648067 AACS | TMEM132B A 0.2449 0.5034 0.3375 0.751 0.0007698 18 RS4306606 55334295 ATP8B1 A 0.2713 0.5341 0.3691 0.773 0.0008836 7 4 RS7656244 65170415 SRD5A2L2 A 0.1859 0.4634 0.2959 0.7255 0.0007708 18 RS2345 75279459 GALR1 | LOC100132713 G 0.06883 2.148 1.368 3.372 0.0008951 2 RS6547192 78434082 SNAR-H | REG3G A 0.1373 1.89 1.304 2.739 0.0007725 3 RS4974500 134109630 AMOTL2 | ANAPC13 G 0.4943 0.6023 0.4466 0.8124 0.0008961 8 2 RS6432964 149986664 LYPD6B G 0.3654 0.5723 0.4133 0.7924 0.0007749 3 RS1879784 183203034 MCF2L2 | KLHL6 A 0.4757 1.643 1.226 2.202 0.000898 8 RS167069 115423582 CSMD3 | TRPS1 G 0.4848 0.6037 0.4497 0.8102 0.0007753 4 RS12511494 113770535 ANK2 A 0.141 1.859 1.289 2.682 0.0009018 9 8 RS10283337 1319724 LOC401442 | DLGAP2 G 0.4052 1.625 1.224 2.157 0.0007815 6 RS2474880 63460047 KHDRBS2 | LOC100128610 A 0.4325 1.638 1.224 2.192 0.0009037 4 RS7676913 118043203 TRAM1L1 | LOC100288955 C 0.4706 0.6035 0.4495 0.8104 0.0007854 15 RS11636337 70044354 LOC145837 | C15orf50 G 0.251 0.5284 0.3625 0.7704 0.0009116 10 1 RS493631 153143048 SPRR2G | LELP1 A 0.1195 1.968 1.325 2.921 0.0007874 11 RS1852755 13996686 SPON1 G 0.2611 0.529 0.3631 0.7707 0.000912 14 RS878554 94606179 IFI27L2 | PPP4R4 A 0.2057 1.746 1.261 2.418 0.0007884 9 RS16933086 36215581 GNE A 0.2016 1.685 1.238 2.293 0.0009121 11 11 RS10894162 129781016 PRDM10 A 0.02092 3.344 1.652 6.767 0.000792 8 RS12547514 55806439 LOC100287651 | XKR4 C 0.4172 1.638 1.224 2.194 0.0009135 5 RS1154758 162716850 GABRG2 | CCNG1 G 0.4443 0.5973 0.4421 0.8071 0.0007932 7 RS13237260 7779091 RPA3 | LOC729852 A 0.2415 1.665 1.232 2.252 0.000918 12 10 RS7914288 71739556 COL13A1 | H2AFY2 G 0.389 1.629 1.225 2.167 0.0007965 13 RS9532283 39394277 FREM2 A 0.4673 1.626 1.22 2.168 0.0009189 17 RS7219555 59740122 NACA2 | BRIP1 C 0.3151 1.636 1.227 2.181 0.0007979 2 RS12619691 35306866 MYADML | LOC100288911 A 0.3937 1.628 1.22 2.172 0.0009197 13 12 RS10783915 59001589 XRCC6BP1 | LRIG3 G 0.332 0.5723 0.413 0.793 0.000798 3 RS6766938 67150565 KBTBD8 | SUCLG2 A 0.2142 1.721 1.248 2.373 0.0009197 5 RS951087 13328289 CTNND2 | DNAH5 A 0.1107 1.971 1.325 2.931 0.0008055 11 RS1852757 13987892 SPON1 G 0.1866 0.4711 0.3018 0.7353 0.0009207 14 13 RS9510573 23680990 LOC646201 | SGCG A 0.1889 1.746 1.26 2.419 0.0008068 9 RS7846713 8079286 C9orf123 | PTPRD A 0.2375 1.7 1.242 2.327 0.0009219 14 RS885845 96729728 BDKRB1 A 0.37 1.639 1.228 2.189 0.0008088 5 RS457638 80359968 RASGRF2 G 0.1867 0.468 0.2986 0.7333 0.0009227 15 1 RS6676869 209249614 PLXNA2 | LOC642587ForG 0.3036peer 0.5568 0.3952 0.7844 0.0008115review19 RS9917042 14494473 CD97 onlyG 0.3212 1.623 1.219 2.162 0.0009241 6 RS9283918 53926867 C6orf142 A 0.05904 2.267 1.404 3.661 0.0008121 8 RS12677088 19309945 CSGALNACT1 A 0.3126 0.5637 0.4015 0.7913 0.0009254 16 6 RS9296737 53926716 C6orf142 G 0.05904 2.267 1.404 3.661 0.0008121 6 RS6570859 148829305 SASH1 G 0.3333 0.5748 0.4141 0.7977 0.0009283 3 RS9841113 116455714 LOC285194 | IGSF11 A 0.4339 1.62 1.222 2.149 0.0008144 15 RS7183427 39503508 LOC100289563 | C15orf54 C 0.1835 1.766 1.261 2.472 0.0009291 17 5 RS1023617 172740329 NKX2-5 | STC2 A 0.3128 1.66 1.234 2.235 0.0008162 10 RS11003908 55645842 PCDH15 A 0.02092 4.64 1.87 11.51 0.0009301 3 RS4076927 46730749 ALS2CL G 0.3171 0.55 0.3875 0.7807 0.0008215 14 RS1197382 81146776 C14orf145 G 0.08974 0.2693 0.1238 0.5856 0.0009328 18 8 RS10091115 62684067 LOC100287835 | NKAIN3 C 0.3381 0.5715 0.4117 0.7934 0.0008288 7 RS3800895 2125942 MAD1L1 A 0.2743 1.636 1.222 2.19 0.0009334 12 RS2364227 93893510 MRPL42 G 0.303 0.5558 0.3939 0.7844 0.0008315 10 RS7905027 124114432 BTBD16 | PLEKHA1 G 0.3932 0.5897 0.4313 0.8063 0.0009347 19 21 RS2832279 30613466 C21orf109 A 0.3323 0.5744 0.415 0.7951 0.0008317 2 RS6748959 175424127 GPR155 | WIPF1 G 0.4906 0.6052 0.4495 0.8148 0.0009351 4 RS12508492 112696927 LOC729065 | C4orf32 G 0.05331 2.324 1.417 3.81 0.0008328 9 RS10814369 36292975 GNE | LOC646993 A 0.192 1.714 1.246 2.359 0.0009351 20 6 RS6940801 169655667 THBS2 | LOC100289536 G 0.1498 1.779 1.269 2.493 0.0008332 11 RS12793784 26700112 SLC5A12 A 0.1208 1.926 1.306 2.84 0.0009361 3 RS17485052 138638081 PIK3CB | FOXL2 A 0.1991 1.75 1.26 2.43 0.0008398 16 RS6564372 73324074 LOC100288121 | LOC401859 A 0.4986 0.6197 0.4668 0.8228 0.0009375 21 12 RS10746042 106111254 C12orf75 | NUAK1 G 0.2399 1.695 1.244 2.311 0.0008398 2 RS7559847 222068646 SLC4A3 | EPHA4 A 0.385 1.639 1.223 2.197 0.0009419 22 RS138154 46107105 ATXN10 G 0.2319 1.673 1.237 2.263 0.0008421 3 RS9790156 139814577 CLSTN2 A 0.1002 1.978 1.32 2.964 0.0009425 22 1 RS12120640 74821837 TNNI3K G 0.1205 1.86 1.292 2.678 0.0008424 1 RS1925421 39250964 LOC400750 | RRAGC A 0.2567 1.645 1.225 2.21 0.0009447 20 RS2425817 44895286 CDH22 | SLC35C2 A 0.4329 1.633 1.225 2.179 0.0008438 23 RS10521479 55583383 LOC644893 | FOXR2 A 0.1163 2.228 1.386 3.582 0.0009464 23 7 RS4730197 106446900 FLJ36031 | PIK3CG C 0.1323 0.3697 0.2061 0.6631 0.0008448 5 RS7446493 6974432 POLS | LOC442132 G 0.1029 1.905 1.3 2.791 0.0009471 6 RS2817457 156943434 NOX3 | ARID1B A 0.4774 1.648 1.229 2.21 0.0008462 1 RS3907135 106526262 LOC642337 | LOC126987 G 0.4926 1.628 1.219 2.173 0.0009488 24 1 RS3006993 114570444 OLFML3 | SYT6 A 0.2696 1.671 1.236 2.258 0.0008466 2 RS4588165 35326722 MYADML | LOC100288911 A 0.414 1.631 1.22 2.18 0.0009507 11 RS7128466 120491438 ARHGEF12 | GRIK4 A 0.1437 0.4015 0.2349 0.6863 0.000849 5 RS2401793 13332878 CTNND2 | DNAH5 C 0.1123 1.952 1.313 2.904 0.0009537 25 11 RS11025673 20663967 SLC6A5 G 0.3812 1.6 1.214 2.109 0.000851 3 RS11709505 53749498 CACNA1D G 0.2328 1.684 1.236 2.294 0.0009593 3 RS1523759 77604242 ROBO2 G 0.1711 0.4267 0.2586 0.7039 0.0008541 2 RS9288433 212407077 ERBB4 A 0.3101 1.634 1.221 2.187 0.000961 26 12 RS10746041 106108279 C12orf75 | NUAK1 G 0.2264 1.711 1.248 2.347 0.0008554 10 RS4751800 122636730 BRWD2 G 0.1559 0.4236 0.2543 0.7054 0.0009626 6 RS4895999 134147153 MGC34034 G 0.3805 1.614 1.218 2.139 0.0008575 23 RS6639393 4052872 LOC729162 | NLGN4X G 0.2684 1.885 1.294 2.747 0.0009636 27 3 RS10804937 103308760 LOC100287880 | ALCAM A 0.3184 1.624 1.221 2.16 0.0008578 20 RS6109068 11588817 JAG1 | LOC728573 G 0.1211 0.3592 0.1955 0.6599 0.0009683 13 RS9557607 101950392 NALCN G 0.03036 2.92 1.555 5.484 0.0008589 7 RS7788775 141817070 LOC93432 A 0.2318 0.5088 0.3406 0.7601 0.0009702 28 10 RS10826503 29010588 BAMBI | LOC729601 A 0.3866 1.636 1.225 2.185 0.0008601 15 RS1426079 70044495 LOC145837 | C15orf50 A 0.4123 1.63 1.219 2.178 0.0009716 10 RS619030 11206281 CUGBP2 A 0.2328 1.712 1.248 2.349 0.0008633 7 RS2613585 111186693 IMMP2L G 0.1164 0.317 0.1601 0.6274 0.0009726 29 18 RS151472 3447871 TGIF1 A 0.2908 1.652 1.229 2.219 0.0008653 9 RS1889321 113300835 SVEP1 G 0.2682 1.65 1.225 2.221 0.0009738 1 RS476093 181787259 CACNA1E | ZNF648 C 0.2446 1.683 1.239 2.285 0.0008663 10 RS7094645 72796318 PCBD1 | UNC5B A 0.2481 0.5231 0.3559 0.7689 0.0009773 30 1 RS12130076 240918105 GREM2 | LOC645939 G 0.1238 0.3612 0.1983 0.6577 0.0008678 3 RS7633016 46728659 ALS2CL A 0.3192 0.5552 0.3913 0.7877 0.0009774 1 RS2169089 5088617 AJAP1 | LOC100287877 A 0.2328 0.5191 0.3529 0.7635 0.0008679 18 RS6506772 75918499 GALR1 | LOC100132713 G 0.254 0.5182 0.3505 0.7662 0.000983 31 20 RS6032678 44777295 CD40 | CDH22 A 0.4514 1.631 1.223 2.176 0.000871 3 RS13084580 39188182 CSRNP1 A 0.1111 1.906 1.299 2.798 0.000987 12 RS4766019 3089354 TEAD4 A 0.123 0.3318 0.1733 0.6353 0.0008719 7 RS10276017 140979971 LOC100129514 | LOC100130169 A 0.04082 2.507 1.451 4.333 0.0009892 32 2 RS16985278 18734072 KCNS3 | RDH14 A 0.2635 1.684 1.239 2.288 0.0008739 13 RS7323413 55363417 LOC100287887 | PRR20 C 0.03745 3.145 1.59 6.221 0.0009912 3 RS1163368 108946466 C3orf66 | DPPA2 A 0.3219 1.621 1.22 2.155 0.0008751 9 RS6415721 14964952 FREM1 | LOC389705 G 0.319 0.571 0.409 0.797 0.0009913 33 7 RS11971933 111717031 DOCK4 G 0.09615 2.053 1.344 3.135 0.0008756 20 RS11086320 49687647 MULTIPLE_GENES:3755,100128598 G 0.3013 1.662 1.228 2.249 0.0009953 18 RS7506330 9177894 ANKRD12 G 0.4774 0.6034 0.4481 0.8125 0.0008765 4 RS1995917 24207765 PPARGC1A | LOC729175 G 0.06545 2.127 1.357 3.333 0.0009991 http://bmjopen.bmj.com/ 34 6 RS9442693 72386001 C6orf155 | RIMS1 A 0.05027 2.582 1.476 4.514 0.0008798 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 24
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 51 BMJ Open
1 C. SUPPLEMENTAL REFERENCES BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 1. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components 4 analysis corrects for stratification in genome wide association studies. Nat Genet. 2006;38:904 5 909 6 2. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by 7 genome wide association. Genomics. 2008;92:265 272 8 9 3. Liu KY, Muehlschlegel JD, Perry TE, Fox AA, Collard CD, Body SC, Shernan SK. Common 10 genetic variants on chromosome 9p21 predict perioperative myocardial injury after coronary 11 artery bypass graft surgery. J Thorac Cardiovasc Surg. 2010;139:483 488, 488 e481 482 12 4. Muehlschlegel JD, Liu KY, Perry TE, Fox AA, Collard CD, Shernan SK, Body SC. Chromosome 13 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation. 14 2010;122:S60 65 15 5. Kraft P. EfficientFor two stage peer genome wide review association designs only based on false positive report 16 probabilities. Pac Symp Biocomput. 2006:523 534 17 6. Zuo Y, Zou G, Wang J, Zhao H, Liang H. Optimal two stage design for case control association 18 analysis incorporating genotyping errors. Ann Hum Genet. 2008;72:375 387 19 20
21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 25
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 51
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 STROBE Statement—checklist of items that should be included in reports of observational studies 3 4 Item 5 No Recommendation 6 Title and abstract 7 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 8 (b) Provide in the abstract an informative and balanced summary of what was done 9 and what was found 10 Introduction 11 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 13 Objectives 3 State specific objectives, including any prespecified hypotheses 14 Methods 15 For peer review only 16 Study design 4 Present key elements of study design early in the paper 17 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 18 exposure, follow-up, and data collection 19 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of 20 21 selection of participants. Describe methods of follow-up 22 Case-control study—Give the eligibility criteria, and the sources and methods of 23 case ascertainment and control selection. Give the rationale for the choice of cases 24 and controls 25 26 Cross-sectional study—Give the eligibility criteria, and the sources and methods of 27 selection of participants 28 (b) Cohort study—For matched studies, give matching criteria and number of 29 exposed and unexposed 30 Case-control study—For matched studies, give matching criteria and the number of 31 32 controls per case 33 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect http://bmjopen.bmj.com/ 34 modifiers. Give diagnostic criteria, if applicable 35 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 36 37 measurement assessment (measurement). Describe comparability of assessment methods if there 38 is more than one group 39 Bias 9 Describe any efforts to address potential sources of bias 40 Study size 10 Explain how the study size was arrived at 41
42 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, on October 1, 2021 by guest. Protected copyright. 43 describe which groupings were chosen and why 44 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 45 (b) Describe any methods used to examine subgroups and interactions 46 (c) Explain how missing data were addressed 47 48 (d) Cohort study—If applicable, explain how loss to follow-up was addressed 49 Case-control study—If applicable, explain how matching of cases and controls was 50 addressed 51 Cross-sectional study—If applicable, describe analytical methods taking account of 52 53 sampling strategy 54 (e) Describe any sensitivity analyses 55 Continued on next page 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 Page 51 of 51 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Results 4 5 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, 6 examined for eligibility, confirmed eligible, included in the study, completing follow-up, and 7 analysed 8 (b) Give reasons for non-participation at each stage 9 (c) Consider use of a flow diagram 10 11 Descriptive 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information 12 data on exposures and potential confounders 13 (b) Indicate number of participants with missing data for each variable of interest 14 (c) Cohort study—Summarise follow-up time (eg, average and total amount) 15 For peer review only 16 Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 17 Case-control study—Report numbers in each exposure category, or summary measures of 18 exposure 19 Cross-sectional study—Report numbers of outcome events or summary measures 20 21 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their 22 precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and 23 why they were included 24 (b) Report category boundaries when continuous variables were categorized 25 26 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful 27 time period 28 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 29 analyses 30 31 Discussion 32 Key results 18 Summarise key results with reference to study objectives 33 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. 34 http://bmjopen.bmj.com/ Discuss both direction and magnitude of any potential bias 35 36 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity 37 of analyses, results from similar studies, and other relevant evidence 38 Generalisability 21 Discuss the generalisability (external validity) of the study results 39 Other information 40 41 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable,
42 for the original study on which the present article is based on October 1, 2021 by guest. Protected copyright. 43 44 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and 45 46 unexposed groups in cohort and cross-sectional studies. 47 48 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 49 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 50 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 51 52 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 53 available at www.strobe-statement.org. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
Genome-wide association study of perioperative myocardial infarction after coronary artery bypass surgery
ForJournal: peerBMJ Open review only Manuscript ID: bmjopen-2014-006920.R1
Article Type: Research
Date Submitted by the Author: 25-Feb-2015
Complete List of Authors: Kertai, Miklos; Duke University, Department of Anesthesiology/Cardiothoracic Li, Yi-Ju; Duke University, Department of Biostatistics and Bioinformatics Li, Yen-Wei; Duke University, Department of Biostatistics and Bioinformatics Ji, Yunqi; Duke University, Department of Biostatistics and Bioinformatics Alexander, John; Duke University, Department of Medicine/Cardiology; Duke University, Duke Clinical Research Institute Newman, Mark; Duke University, Department of Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute Smith, Peter; Duke University, Department of Surgery/Cardiothoracic Joseph, Diane; Duke University, Duke Clinical Research Institute Mathew, Joseph; Duke University, Department of Anesthesiology/Cardiothoracic http://bmjopen.bmj.com/ Podgoreanu, Mihai; Duke University, Department of Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute
Primary Subject Genetics and genomics Heading:
Secondary Subject Heading: Cardiovascular medicine, Surgery, Epidemiology on October 1, 2021 by guest. Protected copyright. Myocardial infarction < CARDIOLOGY, SURGERY, Adult intensive & critical Keywords: care < ANAESTHETICS, Anaesthesia in cardiology < ANAESTHETICS, Cardiac Epidemiology < CARDIOLOGY, GENETICS
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from ; Mark F. Newman, F. Newman, ;Mark 2,6 ; ; 1,6 ; John Alexander, MD, MPH MD, JohnAlexander, ; 4 Department of Biostatistics and Bioinformatics; Bioinformatics; of Biostatisticsand Department 4 ; Mihai V. Podgoreanu, MD Podgoreanu, V. ;Mihai 1 : 3,100
Cardiac Surgery; Surgery; Cardiac 3 BMJ Open , Ph.D.; Yunqi Ji, PhD Yunqi Ph.D.; , http://bmjopen.bmj.com/ 4 ; Joseph P. Mathew, MD, MPH MD, JosephP. Mathew, ; 7 Cardiology, and and Cardiology, ; Yen Wei Li ;Yen Wei 2 4,5 Duke Clinical Research Institute; Duke University, Durham, NC, USA USA NC, Durham, University, Duke Institute; Research Clinical Duke on October 1, 2021 by guest. Protected copyright. 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml ; Yi Ju Li, PhD ;Yi Ju Li, ; Diane Joseph, MA ;Diane 1 3 For peer review only cardiopulmonary bypass, genome wide association study, inflammation, myocardial infarction, surgery infarction, inflammation, myocardial study, association bypass, genome wide cardiopulmonary ; Peter K. Smith, MD Smith, K. ;Peter 1,6 Divisions of Cardiothoracic Anesthesiology, Anesthesiology, Cardiothoracic of Divisions Physiology Institute; Molecular Duke Duke University Medical Center, Durham, NC 27710. 27710. NC Durham, Center, Medical University Duke Email: [email protected] 919 681 4720; Telephone: 919 681 4776; Fax: Words: Key for the Duke Perioperative Genetics and Safety Outcomes (PEGASUS) Investigative Team Investigative (PEGASUS) Outcomes Safety and Genetics Perioperative Duke the for MD Genome-wide association study of perioperative myocardial infarction after coronary artery bypass surgery bypassartery surgery coronary after infarction myocardial perioperative study of association Genome-wide PhD MD, Kertai, MiklosD. 1 5 of Anesthesiology, MD; Department 3094, Box MihaiPodgoreanu, to V. Correspondence references) and tables, figures abstract, title page, (excluding Count Word Page 1 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 2 of 52 in stage II. A secondary secondary A instage II. 5 2 for rs609418), or a dichotomized phenotype based on CK on extreme based phenotype dichotomized a or rs609418), for 3 BMJ Open http://bmjopen.bmj.com/ The primary outcome variable was perioperative MI, defined as CK MB values CK MB as MI, defined perioperative variable was outcome The primary on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We conducted a stage I genome wide association study (GWAS) in 1,433 ethnically diverse patients of both genders both of genders patients diverse ethnically (GWAS)in1,433 study association stage a genome wide I conducted We Identification of patient populations at risk for developing myocardial infarction (MI), or alternatively, with intrinsic with or (MI), alternatively, infarction myocardial developing at for populations risk patient of Identification Following quality control and adjustment for clinical covariates, we identified 521 single nucleotide polymorphisms (SNPs) polymorphisms nucleotide single 521 weidentified covariates, clinical for and adjustment control quality Following ABSTRACT ABSTRACT Objectives:
analysis using CK MB as a quantitative trait (min p = 1.26x10 = p (min trait quantitative as CK MB a using analysis Participants: cardioprotection, or responsiveness to a protective agent remain high priority knowledge gaps. We sought to identify novel common novel toidentify sought We gaps. knowledge high priority remain agent protective toa responsiveness or cardioprotection, using genome wide bypass grafting artery coronary undergoing MI patients in perioperative with associated variants genetic methodology. association States. the United centers across surgery cardiac and tertiary secondary 107 Setting: ≥10xULN during the first postoperative day, and not attributable to preoperative MI. Secondary outcomes included postoperative CK postoperative included outcomes Secondary MI. preoperative to and attributable not day, first during the postoperative ≥10xULN distribution. CK MB the of quartiles extreme on based phenotype dichotomized a or trait, quantitative a as MB Results: (112 cases/1,321 controls) from the Genetics of Myocardial Adverse Outcomes and Graft Failure (GeneMAGIC) study, and a stage study, a stage II and (GeneMAGIC) Failure Graft and Adverse Outcomes Myocardial of Genetics the from controls) cases/1,321 (112 Duke and the GeneMAGIC from combined controls) cases/1,830 (225 patients 2,055 of inan expanded population analysis bypass coronary non emergent primary undergoing Patients studies. (PEGASUS) Outcomes and Safety Genetics Perioperative included. were grafting measures: and outcome secondary Primary in the stage I GWAS analysis. Among these, 8 common variants in 3 genes or intergenic regions met p < 10 < p regionsmet intergenic or genes in3 common these, 8 variants Among analysis. GWAS stage I inthe 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 3 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml for rs4834703) supported these findings. Pathway analysis revealed that genes harboring top scoring scoring top harboring that genes revealed analysis Pathway findings. these supported rs4834703) for 6 For peer review only Using 2 stage GWAS and pathway analysis, we identified and prioritized several potential susceptibility loci for loci potential susceptibility several and prioritized we identified analysis, pathway and GWAS 2 stage Using We conducted comprehensive and complementary single marker and pathway based genome wide association analyses. analyses. association pathway based and genome wide marker single complementary and comprehensive conducted We This is the first genome wide association study of perioperative myocardial infarction, using prospective cohorts of cardiac cardiac of cohorts prospective using myocardial of infarction, study perioperative association genome wide first is the This factors. risk non genetic and adjustment full phenotype, for primary the of definitions standard patients, surgical The study is powered to detect relatively large effect sizes. large effect relatively to is powered detect study The Predominantly Caucasian cohort, thus findings cannot be generalized to other populations. populations. toother cannot generalized be findings thus Caucasian cohort, Predominantly Rare genetic variant effects not analyzed. analyzed. not effects variant genetic Rare
• • • • • MB values (min p = 7.72x10 = p (min values MB
variants cluster in pathways of biologic relevance to extracellular matrix remodeling, ER to Golgi transport and inflammation. and transport ER to Golgi matrix remodeling, toextracellular of relevance biologic pathways in cluster variants Conclusions: perioperative MI. perioperative STUDY THIS OF LIMITATIONS AND STRENGTHS
Page 3 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 4 of 52
1 Thus, the overall influence of of commoninfluence overall the Thus, 8 12
2 4 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. However, although family based methods are not practical for studying perioperative MI, its genetic MI, its genetic perioperative for studying practical not methods are family based although However, 3 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Yet only a few studies have identified allelic associations with altered susceptibility to myocardial ischemia reperfusion ischemia reperfusion tomyocardial susceptibility altered with allelic associations identified few have a only studies Yet 7 The evidence for heritability of MI is striking, supported both by family studies and more recently, by a number of well of number a by recently, more studies and family by both supported ofis striking, MI heritability for evidence The Despite advances in surgical techniques and pharmacologic therapy, the incidence of myocardial infarction (MI) after coronary coronary after (MI) infarction of myocardial the incidence therapy, pharmacologic and techniques insurgical advances Despite individual genetic background may also significantly modulate the magnitude of postoperative inflammatory response after cardiac after cardiac response inflammatory postoperative magnitudeof the modulate significantly also backgroundmay genetic individual surgery. genetic variation on the incidence of perioperative MI remains poorly understood. understood. poorly remains MI of perioperative incidence the on variation genetic Identification of populations at risk for developing large infarcts or, alternatively, with predictors of a cardioprotective state, state, or cardioprotective of a predictors with alternatively, or, infarcts large developing for populations risk at of Identification knowledge gaps. high priority remain protective agent toa responsiveness clinical and procedural risk factors, different racial susceptibility profiles, and results from preclinical animal models. Indeed, models. Indeed, animal preclinical from and profiles, results different racial susceptibility factors, risk procedural and clinical the such as MI, postoperative of pathophysiology implicated inthe pathways biological in found been variability has genetic extensive that suggests state pro inflammatory of a heritability for evidence Mounting surgery. tocardiac response inflammatory acute complex injury in cardiac surgical populations, all based on a candidate gene approach. association candidate gene a on based surgical all populations, incardiac injury INTRODUCTION INTRODUCTION
locus in multiple racial groups. inmultiple racial locus artery bypass grafting (CABG) remains as high as 19%, and is associated with increased mortality and long term morbidity. long term mortality and increased with is associated and high 19%, as as remains (CABG) bypass grafting artery basis is strongly suggested by several observations, including wide variability in incidence and severity that is poorly explained by by explained is poorly that severity and inincidence variability wide observations, including several by suggested is strongly basis powered and replicated genome wide association studies (GWAS), which primarily implicate common genetic variants at the 9p21 9p21 the at variants genetic implicate common primarily which (GWAS), studies association replicated genome wide and powered 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Given the polygenic polygenic the Given 13 15 5 Of those, 1,433 patients met eligibility requirements after applying applying after requirements eligibility met patients 1,433 those, Of 18 BMJ Open http://bmjopen.bmj.com/ genome wide association study combined with a pathway analysis approach. analysis approach. pathway a with combined study association genome wide 17 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We performed joint two stage a performed We 16 The stage I cohort (discovery cohort) was comprised of 1,493 ethnically diverse subjects who underwent an isolated CABG CABG isolated an underwent who subjects diverse ethnically 1,493 of comprised was cohort) (discovery cohort stage I The Recently, integrated testing of genes involved in the same biological pathway has emerged as an alternative strategy for strategy alternative as an has emerged pathway biological inthe same involved of testing genes integrated Recently,
nature of disease susceptibility, this approach is increasingly being used to identify groups of gene variants with shared cellular shared with variants of toidentify gene groups used being increasingly is this approach susceptibility, disease of nature by this strategy, we adopted this study GWAS. of power In statistical the improving also while disease, enrichedfor are that function after MI perioperative with associated variants common genetic toidentify methodology association genome wide employing first quality control criteria and excluding patients with missing genotypes or phenotypic information. phenotypic or genotypes missing with patients criteria andexcluding control quality with cardiopulmonary bypass for the first time, and were enrolled between 2002 and 2003 in the Institutional Review Board (IRB) Board Review inthe 2003 Institutional and between 2002 enrolled and were time, first the bypass for cardiopulmonary with Ex Project of of multicenter the sub study ancillary (GeneMAGIC) Failure and Outcomes Graft Myocardial of Genetics approved,
populations Patient CABG, followed by pathway based analyses to uncover biological mechanisms of relevance to perioperative MI. toperioperative mechanisms relevance biological of touncover pathway based analyses by followed CABG, vivo Vein Graft Engineering via Transfection (PREVENT IV). Transfection via Engineering Graft Vein vivo METHODS METHODS Association Genetic of the Reporting “Strengthening by recommendations the follows the of results reporting design and study The (STREGA). Studies” evaluating the combined effects of multiple genetic variants with small effect size on a disease phenotype. disease a size effect small on with variants multiplegenetic of effects combined the evaluating Page 5 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 6 of 52 see as an elevation in plasma level of creatine kinase kinase creatine of level plasma in elevation an as 19 method to control for population stratification ( population stratification for tocontrol method 21
6 11 12 BMJ Open http://bmjopen.bmj.com/
18 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We used principal components derived from EIGENSTRAT from derived components principal used We 20 ancestry were added to the discovery dataset leading to a total of 2,055 patients. The additional patients underwent CABG with with CABG patientsunderwent additional patients. The 2,055 of total toa leading dataset tothe discovery were added ancestry an (PEGASUS), IRB Outcomes Study Safety and Genetics Perioperative the part of as 1997 2006 bypass between cardiopulmonary Center. Medical University Duke at study longitudinal approved on the Illumina Human610 Quad BeadChip at the Duke Genomic Analysis Facility. Sample and genotype quality control of data flow of data flow control quality and genotype Sample Analysis Facility. Genomic the at Duke BeadChip Human610 Quad the on Illumina previously as equilibrium, Hardy Weinberg and SNP missingness, relatedness, cryptic check, call gender rates, of assessment included described. Supplemental Methods ). Supplemental MB isoenzyme (CK MB) that was > 10 times the upper limit of normal, as measured by a core laboratory within 24 hours after after hours 24 within laboratory by core a as measured limitnormal, of the times 10 upper > was that (CK MB) isoenzyme MB European self reported of patients 622 which additional dataset, an in expanded variants the genetic top tested stage we II, In infarction myocardial perioperative Definitionof MI of definition universal tothe defined MI according was Perioperative controls and quality Genotyping performed was cohorts both in Genotyping procedures. standard using or blood saliva whole was isolated from DNA Genomic surgery, and that was not attributable to an intervening clinical event or pre operative myocardial infarction (adjudicated by the by (adjudicated infarction myocardial pre operative or event clinical intervening toan not that and was attributable surgery, Events Clinical Committee). PREVENT IV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from arbitrarily arbitrarily quartiles of the CK MB CK MB the of quartiles CK MB quartile, and “1” for for “1” and quartile, CK MB th st and 4 and st 7 a priori a was analyses for stage I significance Statistical ). BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. see Supplemental TableS1 Supplemental see For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only CK MB quartile from the stage II cohort. Logistic regression with the same set of covariates was performed. performed. was set covariates of same the with regression stage Logistic cohort. the II from quartile CK MB th defined replication dataset to obviate type I error. In stage II analyses, the same clinical covariate and principal component principal and clinical covariate the same stage analyses, II In type error. toobviate I dataset replication defined Univariate regression analysis was performed to test differences in demographic, clinical and procedural characteristics characteristics procedural and clinical indemographic, totest differences performed analysis was regression Univariate a priori a defined as a 2 tailed p < 0.001, to balance between the overly conservative Bonferroni correction and type II error, given that we had had that we error, given II andtype correction Bonferroni conservative overly the between balance to < p 0.001, 2 tailed as a defined an adjust for population stratification ( stratification population for adjust adjustments (re estimated for the expanded dataset) were applied, with statistical significance defined as a Bonferroni adjusted p value adjustedp value Bonferroni a defined as significance statistical with applied, dataset) were expanded the for (re estimated adjustments PLINK 1.07, assuming an additive genetic model, and including significant clinical covariates and top 10 principal components to components top principal 10 and covariates significantclinical including model, and genetic 1.07, an additive assuming PLINK of 0.05/number of SNPs identified in stage I, to control the overall type I error rate. Haplotype association analysis was performed in was performed association analysis Haplotype rate. type error I overall tocontrolthe instage I, SNPs identified of 0.05/number of subjects within the 4 the within subjects Genome wide association analyses performed in the stage I cohort used multivariable logistic regression models implemented in implemented models regression logistic multivariable used inthe cohort I stage performed analyses association Genome wide distribution. For CK MB, a linear regression model was applied with adjustment for the same set of covariates as in the primary primary inthe as of set covariates same the adjustment with for applied model was regression linear a CK MB, For distribution. 1 the within subjects for “0” trait as status we re assigned subset, CK MB the extreme For analysis. analysis Statistical
the stage II cohort for genes tagged by the significant SNPs, adjusting for the same clinical covariates and principal components. components. principal and clinical covariates the same for SNPs, adjusting significant the by tagged for genes cohort stage II the CK MB of the phenotype continuous using cohort, stage II SNPs top in the the identified for were performed analyses Secondary 1 subjects the within whichincluded phenotype, CK MB dichotomized a extreme and values between patients with and without postoperative MI, and when statistically significant were used to adjust genetic association tests. tests. association used to genetic were adjust significant statistically when and MI, postoperative without patients and with between Page 7 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 8 of 52 ), 521 521 ), Figure 1 Figure ). ). ), 534,390 markers were were markers 534,390 ), ), were in high linkage inhigh were ), for 521 SNPs). The top 2 SNPs SNPs 2 top The SNPs). 521 for 5 (OR 0.46; 95% CI, 0.33 0.63), 0.33 0.63), CI, 95% 0.46; (OR 6 PAPPA2 Supplemental TableS2 Supplemental , rs2044061 on chromosome 8), and were and were 8), chromosome on rs2044061 ,
6 ). ). 8 BMJ Open ). A quantile quantile (Q Q) plot of observed vs expected p values expected vs observed plotof (Q Q) quantile quantile A ). http://bmjopen.bmj.com/ Supplemental Figure S2 Figure Supplemental (OR 0.46; 95% CI, 0.33 0.63) and 2.47x10 and 0.33 0.63) CI, 95% 0.46; (OR 6 ). Aortic cross clamp time, number of coronary artery grafts, and procedures in procedures addition and grafts, artery coronary of number time, cross clamp Aortic ). Supplemental Methods: A.1. Quality control of flow data Quality control A.1. SupplementalMethods: on October 1, 2021 by guest. Protected copyright. Supplemental TableS10 Supplemental For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Supplemental TableS2 Supplemental defined discovery threshold of p< 0.001 (minimum p = 2.76x10 = p (minimum 0.001 p< threshold of discovery defined a priori a After applying quality control criteria ( control criteria quality applying After In stage II analyses, 8 of the 521 SNPs met the Bonferroni correction threshold (p < 9.6 x 9.6 < 10 threshold (p correction of 8 Bonferroni the SNPs met 521 stage the analyses, II In Perioperative MI was observed in 112 of 1,433 patients (7.8%) and in 225 of 2,055 patients (10.9%) in stage I and stage stage and II stage inI (10.9%) patients in225 2,055 of and patients 1,433 of (7.8%) in112 observed MI was Perioperative SNPs met the the SNPsmet analyzed for association with perioperative MI in stage I. While none of the SNPs reached genome wide significance ( significance the genome wide of SNPs reached While none MI I. instage perioperative with association for analyzed disequilibrium, with p values for stage II of 2.43x10 of stage for II p values with disequilibrium, to CABG were significantly associated with perioperative MI in stage I analyses. Stage II analyses showed that extracardiac extracardiac that Stage showed analyses II analyses. instage MI I perioperative with associated significantly were toCABG ( potential confounders for tests association genetic the toadjust models regression logistic
(rs10454444 and rs10913237), located in the pregnancy associated plasma protein A2 gene ( A2 gene plasma protein pregnancy associated inthe located rs10913237), and (rs10454444 cohorts, respectively ( respectively cohorts, arteriopathy and year of surgery also played significant roles. All clinical variables identified above were included in multivariable inmultivariable included above were identified variables clinical All roles. significant played also surgery of and arteriopathy year subsequently analyzed in stage ( instage II analyzed subsequently ( for adjusted was well substructure population that showed Finally, we employed pathway analysis to prioritize association results and provide biological interpretation, using functional ontology ontology functional using interpretation, biological and provide results association prioritize to pathway analysis employed we Finally, ). Methods see Supplemental St. Joseph, MI; (GeneGO, toolsinMetaCore analysis implemented enrichment RESULTS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
-6
-6 -6 -6 -6 -6 -6 -6 ). ). 8.89 x10 2.47 x10 2.43 x10 8.18 x10 6.80 x10 9.55 x10 5.81 x10 4.70x10
a member of of member a
, SEC24 related SEC24 related ,
Table1 0.76) ataset* - d 0.63) 0.63) 3.17) 2.66) 2.05) 2.09) 0.65)
------SEC24D, SEC24D, SEC24D (1.32 OR (95%OR CI) P value 0.62 (0.50 0.46 (0.33 0.46 (0.33 2.23 (1.57 1.98 (1.47 1.65 1.67 (1.34 0.47 (0.34
Stage II: Expanded 0.11 0.11 0.11 0.16 0.36 0.36 0.33 0.11 Cases N=1,830
MAF
0.19 0.19 0.06 0.09 0.28 0.48 0.25 0.18 N=225
-4 -4 -4 -5 -4 -4 -4 -4 ), 2 in SEC24 family, member D ( D member inSEC24family, 2 ),
2.67x10 , regulatory factor X associated protein; protein; X associated factor regulatory , 1.81 x10 4.34 x10 3.13 x10 2.59 x10 3.51 x10 3.14 x10 1.85 x10
HDAC4 9 RFXAP 0.64) 3.46) 3.34) 2.33) 0.78) 2.33) 0.64) 0.66) ------ataset*
BMJ Open (1.42 http://bmjopen.bmj.com/ OR (95%OR CI) P value Controls 0.39 (0.24 2.22 2.27 (1.54 1.73 (1.29 0.58 (0.43 1.73 (1.28 0.39 (0.24 0.41 (0.25 , thyrotropin releasing hormone degrading ectoenzyme; intergenic region is expressed expressed is region intergenic ectoenzyme; hormone degrading thyrotropin releasing ,
0.09 0.13 0.18 0.40 0.35 0.35 0.09 0.09 Stage I: Discovery d Cases
TRHDE N=1,321 MAF
on October 1, 2021 by guest. Protected copyright. 0.18 0.06 0.09 0.29 0.48 0.25 0.18 0.18 N=112
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
, pregnancy associated plasma protein A2; A2; protein plasma pregnancy associated , LOC552889 SMAD9
| |
For peer review only PAPPA2 HDAC4 SEC24D SEC24D PAPPA2 PAPPA2 TRHDE RFXAP PAPPA2
176649181 240234901 119691624 119710421 74471340 37417427 176630416 176566350 , mother against decapentaplegic homolog 9; homolog decapentaplegic against mother ,
237 respectively. In addition, one intronic SNP in histone deacetylase 4 ( deacetylase 4 SNP inhistone intronic one addition, In respectively.
the cytoplasmic Coat Protein II complex), and 2 located in intergenic regions were associated with perioperative MI MI ( perioperative with associated regionswere inintergenic located 2 and complex), Protein Coat II cytoplasmic the SMAD9 Top 8 Single Nucleotide Polymorphisms Associated With Postoperative Myocardial Myocardial Infarction With Postoperative Associated Polymorphisms Nucleotide Single 8 Top , histone deacetylase 4; deacetylase 4; histone , rs10454444 rs10200850 rs4834703 rs6822035 rs2303970 rs609418 rs10913 rs10489478
1 2 4 4 12 13 1 1 Chr SNP Base pair Gene symbol Controls Chr, Chromosome; MAF, minor allele frequency (based on the discovery dataset); OR, odds ratio; 95% CI, 95% confidence interval interval 95%confidence CI, 95% ratio; odds OR, dataset); discovery the on (based frequency allele minor MAF, Chromosome; Chr, Table1. HDAC4 by two franking genes with ‘|’ in between. in between. ‘|’ with genes franking two by protein D; D; protein Page 9 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 10 of 52
*
P value 0.1 0.09 0.08 0.002 0.1 SEC24D
) ) ) ) ) 1.052 1.043 2.326 2.654 1.047 - - - - - 0.578 0.575 0.950 1.242 0.574 0.78 ( 0.77 ( 1.49 ( 1.82 ( 0.77 (
) showed the strongest association, association, strongest the showed ) -4 -5 -6 -4 -4 as CK MB an extreme phenotype 5 * 1.21x10 2.27x10 8.26 x10 7.72 x10 1.30 x10 ). Overall, the rs4834703 SNP rs4834703 in the Overall, ). (p = 8.26x10 = (p Table2
5.59 5.58 5.40 9.26 8.54 - - -
HDAC4 CK MB asCK MB a quantitative trait 10 Β coefficient P value (95% OR CI) BMJ Open http://bmjopen.bmj.com/ ). When the dichotomized extreme CK MB phenotype was studied inthe was phenotype CK MB extreme dichotomized the When ).
18 07 10 18 17 0. 0. 0. 0. 0. (rs4834703 and rs6822035, p=0.002) and rs609418 in the intergenic region intergenic rs609418in the and p=0.002) rs6822035, and (rs4834703 Table2 AF (Overall) M ) and rs10200850 in rs10200850 and ) 6 SEC24D on October 1, 2021 by guest. Protected copyright. (p = 0.0001) 0.0001) ( = (p (p = 7.72x10 = (p
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
PAPPA2 PAPPA2 HDAC4 SEC24D PAPPA2 For peer review onlyPAPPA2 (p=0.001) remained nominally significant ( significant nominally remained (p=0.001) SMAD9
and and 176649181 240234901 119691624 176630416 176566350 RFXAP When CK MB was tested as a quantitative trait, all 8 SNPs remained significantly associated with plasma levels of CK MB. CK MB. of levels plasma with associated significantly all SNPstrait, 8 remained quantitative a tested as was CK MB When
followed by the 2 SNPs 2 in the by followed However, rs4834703 in SEC24D rs4834703 However, stage II cohort, only the in only cohort, SNPs 2 located stage II
consistently showed strong association signals across postoperative MI, quantitative CK MB, and the extreme CK MB traits. CK MB the extreme and CK MB, quantitative MI, postoperative across signals association strong showed consistently between between
rs10454444 rs10200850 rs4834703 rs10913237 rs10489478
1 2 4 1 1 Chr SNP Base pair Gene symbol *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of surgery, of surgery, year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted time. cross clamp aortic and vessels, of diseased number Table2. inthe Phenotype Extreme an and as Trait Quantitative as a CK MB with Polymorphisms Nucleotide Single 8 Top the of Association Joint Analysis Dataset Dataset JointAnalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
). ).
are shown in shown are 0.001 0.002 0.12 , SEC24 related SEC24 related , . Haplotype Haplotype . P
) ) ) SEC24D SEC24D 2.034 1.903 1.047 - - - , and , 1.189 1.152 0.669 1.56 ( 1.48 ( 0.84 ( HDAC4 , ,
OR (95%OR CI) -4 -4
0.01 PAPPA2 4.81 x10 2.26 x10 , OR 1.87; 95% CI, 1.46 2.39, Table3 1.46 2.39, CI, 95% 1.87; OR , 7
, regulatory factor X associated protein; protein; X associated factor regulatory , 2.93 4.42 4.80 -
11 RFXAP BMJ Open Haplotype Frequency http://bmjopen.bmj.com/ gene and results of association tests with incidence of perioperative incidence with tests association of and results gene
, thyrotropin releasing hormone degrading ectoenzyme ectoenzyme hormone degrading thyrotropin releasing , 30 47 26 0. 0. 0. SEC24D TRHDE
on October 1, 2021 by guest. Protected copyright.
LOC552889 SMAD9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml | |
Table 3 and Supplemental TableS3 shown Supplemental and Table3 in are analysis haplotype of results the and , , pregnancy associated plasma protein A2; A2; protein plasma pregnancy associated , SNP2 SNP2 Haplotype SEC24D TRHDE RFXAP For peer review only
PAPPA2 119710421 74471340 37417427 had the most significant association with postoperative MI (p (p MI = 5.54x10 with postoperative association most significant the had Estimated haplotype frequencies in the inthe frequencies haplotype Estimated SNP1 SNP1 The haplotype structures of genic regions surrounding the significant SNPs significant in the regionssurrounding genic structures of haplotype The , mother against decapentaplegic homolog 9; homolog decapentaplegic against mother ,
analysis performed for the LD blocks containing the significant markers showed that the A A haplotype (rs6822035, rs10518325) in rs10518325) (rs6822035, haplotype A A the that showed markers the significant containing blocks the for performed LD analysis SEC24D myocardial infarction in the stage II analysis cohort (n = 2055) = (n cohort analysis stage II inthe infarction myocardial Supplemental Figure S3 Figure Supplemental
Table3. SMAD9 rs6822035 rs2303970 rs609418 , histone deacetylase 4; deacetylase 4; histone ,
4 12 13 Chr, Chromosome; MAF, minor allele frequency (based on the expanded dataset); OR, odds ratio; 95% CI, 95% confidence interval interval 95%confidence CI, 95% ratio; odds OR, dataset); expanded the on (based frequency allele minor MAF, Chromosome; Chr, *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of surgery, of surgery, year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted 95%confidence Ratio, (Odds regression logistic or (Β coefficient) regression linear either timeusing cross clamp aortic and vessels, of diseased number analyses. interval) HDAC4 protein D; D; protein Page 11 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 12 of 52
10 10
× × 0.225 0.509 0.657 0.378 . The most significant most significant The . 8.84 5.54
1.19) 1.31) 1.70) 2.90) - - - - Reference Reference Reference 0.76 (0.48 0.87 (0.58 0.86 (0.43 1.39 (0.67 ).
0.685 0.8592 0.7126 0.0528 0.2874 1.65 (1.33-2.06) 0.2358 1.87 (1.46-2.39) 0.06231 0.07845 0.02636 PMI (n=1830)PMI Patients without
12
Supplemental Figure S4 Figure Supplemental
BMJ Open http://bmjopen.bmj.com/ 0.5927 0.9248 0.6044 0.3956 0.3631 (n=225) 0.03097 0.04425 0.03516 0.009083 cohorts ( cohorts Patients with PMI 8 9
A C A G G G G ------A C A A G C G on October 1, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml and cardiac surgical cardiac and 6 For peer review only rs6822035 rs10518325 rs11098451 rs11098451 rs11098451 rs10518325 rs10518325
rs6828577 rs6822035 A-A rs6822035 rs6828577 rs6822035 rs6822035 rs6822035 rs10518325 A-A rs10518325 rs10518325 rs10518325 S5 Figure shown SNPs inare Supplemental the analysis stage I for maps pathway enriched 10 top The were “Cell adhesion: extracellular matrix remodeling” and “Cytoskeleton remodeling: TGF, WNT, and cytoskeletal remodeling” remodeling” WNT, andcytoskeletal TGF, remodeling: “Cytoskeleton and matrix remodeling” extracellular adhesion: “Cell were OR, odds ratio; 95% CI, 95% confidence interval; SNP, single nucleotide polymorphism; PMI, perioperative myocardial infarction. infarction. myocardial perioperative PMI, polymorphism; nucleotide single SNP, interval; 95%confidence CI, 95% ratio; odds OR, and infarction myocardial associated with SNP previously rs10116277 the including MI, incident postoperative and locus 9p21 the at variation genetic between for association no evidence found we analyses, and II stage I factor adjusted risk multivariate In non surgical both in mortality
*Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted time. cross clamp aortic vessels, of diseased number surgery, 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from SNPs in ) ) in single 6 ) ) and 2 intergenic SEC24D (p = 2.43x10 , SEC24D PAPPA2 HDAC4 , , ) ) or as an extreme CK MB phenotype 6 PAPPA2 ) ) in haplotype analyses, and by 7
13 BMJ Open http://bmjopen.bmj.com/ is an integral component of cytoplasmic coat protein complex II (COPII) showed the most consistent strong association with all three phenotypes SEC24D SEC24D , respectively). The most significant canonical pathway maps in stage I and II comparative comparative and II instage I pathway maps canonical most significant The respectively). , 6 on October 1, 2021 by guest. Protected copyright. ). S5 ). TablesS4 and Supplemental A A (rs6822035, rs10518325, p = 5.54x10 and 7.3x10 and For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 8 For peer review only SEC24D , respectively; respectively; , 3 and 1.1x10 and 4 Among Among these 8 SNPs, rs4834703 in We We present the first report of a 2 stage GWAS involving 225 perioperative MI cases and 1,830 controls. After accounting for regions. regions. The most significant association with perioperative MI was exhibited by rs10454444 in marker marker analyses, and secondary secondary analyses when CK MB was evaluated as a quantitative trait (rs4834703, p=2.43x10 clinical clinical and procedural covariates, we identified 8 significant SNPs mapped to 3 genes ( (rs4834703 and rs6822035, p=0.002). p=0.002). rs6822035, and (rs4834703 (enrichment p value = 1.9x10 = p value (enrichment DISCUSSION evaluated evaluated in this study. The transport protein transport machinery, a key player in vesicle trafficking of secretory proteins from the apparatus endoplasmic for delivery to downstream reticulum compartments. COPII is responsible (ER) for cargo sorting and to vesicle morphogenesis, with roles the in Golgi modulating ER exit, cell surface transport, lipid secretion and cholesterol biosynthesis, and function of G protein coupled receptors. Conditions of ischemia, oxidative injury, or acute phase response result in ER stress through accumulation of misfolded proteins, analyses were “Immune response: NFAT signaling and leukocyte interaction” and “Cell adhesion: extracellular matrix remodeling” (p matrix remodeling” extracellular adhesion: “Cell interaction” and leukocyte and signaling NFAT response: “Immune were analyses 7.2x10 = Page 13 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 14 of 52 and IGF 2, limiting its 2 Supplemental Supplemental Figure S6 and with SEC24D being the only isoform implicated in 22 23 involve characteristic skeletal malformations, whereas its
14 SEC24D BMJ Open The related protein PAPPA1 cleaves IGFBP 4 and is activated and http://bmjopen.bmj.com/ ; however, the role of PAPPA2 in cardiovascular biology has not been However, there are no reports of human diseases associated with genetic 28 3 26 25
27 on October 1, 2021 by guest. Protected copyright. ), ), a member of the HDAC family, mediates changes in chromatin structure by removing acetyl Animal models of defects in HDAC4 24 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only ). . . SEC24D groups groups from the core histones, resulting in transcriptional repression. HDAC4 is highly expressed in myocardium, where it plays an Supplemental TableS5 Supplemental Histone deacetylase 4 ( of of variants in other IGF system component genes, namely IGF 2 and the IGF 1 receptor ( released released from vulnerable atherosclerotic plaques. PAPPA1 has been extensively studied as a cardiovascular risk biomarker for both diagnosis and prognosis of acute coronary syndrome extracellular extracellular matrix secretion. cardioprotective cardioprotective effects following ischemia reperfusion. complete complete disruption results in early embryonic lethality. variation in variation plasma Pregnancy associated protein A2 (PAPPA2) is a metalloproteinase that regulates local insulin like growth factor (IGF) which which leads to activation of the unfolded protein response (UPR) signaling pathway. If the protective mechanisms activated by the UPR are insufficient, cells die by apoptosis and autophagy. However, altered expression or function of SEC24 proteins could also explain ER trapping of misfolded proteins under conditions of ER stress, inhibits both myocardial IGF 1, with implications for inflammation linked angiogenesis and repair processes, bioavailability bioavailability by specifically cleaving IGF binding protein 5 (IGFBP 5). In experimental models of myocardial ischemia, IGFBP 5 previously previously reported. Consistent with our single marker analysis findings, GWAS pathway analysis identified significant contributions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from by 29 TGF β plays importantTGF β plays and complex roles 31
15 ). ). Although no direct functional roles are currently attributed to this BMJ Open http://bmjopen.bmj.com/ suggesting functional roles for HDAC4 in modulating perioperative myonecrosis, and their Supplemental Supplemental Table S7 30 on October 1, 2021 by guest. Protected copyright. , , 3 ). ). For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Consistent with this single locus gene association result, the bone morphogenic protein (BMP) For peer review 32 only ) ) gene, part of the transforming growth factor beta (TGF β) signaling pathway, which is markedly activated in SMAD9 Experimental HDAC inhibition is associated with a profound reduction in ischemia induced myocardial cell death, 29 Supplemental Figure S7 Figure Supplemental ischemia reperfusion ischemia reperfusion injury. intergenic intergenic region, a query of the Regulome and Haploreg databases reports that rs609418 is located within active regulatory elements (GATA2 transcription factor binding site by ENCODE ChIP seq and altering regulatory motifs in Gfi1 and Mef2 by Position Weight Matrix, triggering triggering preconditioning effects and promoting myocardial repair. Genetic variants in HDACs are important determinants of GWAS pathway analysis (p = 5.2x10 One of the intergenic SNPs associated with perioperative MI (rs609418) is located near the mother against decapentaplegic important important role in the regulation of gene expression and apoptosis. is involved in myocardial cell cycle progression, differentiation, and in regulating post infarction inflammatory responses. In animal models, associated signaling with resolution of inflammation, through repression of cytokine and chemokine the gene synthesis, and protection against SMAD myocardial transcription factors is susceptibility susceptibility to cardiovascular diseases, homolog homolog 9 ( infarcted infarcted myocardium. Members of the TGF β superfamily transduce their signal from the membrane to the nucleus via a distinct receptors combination of transmembrane and downstream effectors – the SMAD proteins. pathway, pathway, which also transduces its signals via a SMAD9 dependent cascade, was identified as one of the top scoring pathways in our potential use in predicting individual patient responsiveness to HDAC inhibition. inhibition. toHDAC responsiveness patient individual predicting in use potential Page 15 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 16 of 52
33 This discrepancy between convincing 34
16 Many different reasons may account for this, including inadequate 8 9 BMJ Open http://bmjopen.bmj.com/ Taken together, in subjects with established coronary artery disease such as 5 36 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml as well as meta analyses. For peer35 review only Of note, most of the genes identified through pathway enrichment analysis in this study encode targets for therapeutic 14 locus locus and risk for perioperative MI or mortality after CABG. Indeed, Indeed, our observation is consistent with other studies showing that, although genetic variants at the 9p21 locus are associated with incident coronary artery disease, they may not be associated with the actual risk of MI. sample size, variation in study design, differences in allele frequencies, or variability in the definition of perioperative MI phenotype. MI perioperative definitionof inthe or variability frequencies, inallele differences design, instudy size, variation sample associations associations of 9p21 with greater burden of CAD but not with MI in the presence of underlying CAD has been further confirmed by nested case control studies Surprisingly, we have been unable to replicate previously reported associations between common genetic variants at the 9p21 those those included in our study, any lack of association of the 9p21 locus with subsequent MI could have resulted from the presence of coronary artery disease in both carriers and non carriers of the risk variants at the 9p21 locus, which seems to primarily mediate an phenotype. atherosclerotic Strengths of our study are 1) a relatively large population of cardiac surgery patients, 2) a prospective cohort design, and 3) a approach, replication of association findings at the gene and pathway level is much easier compared to replication at the individual SNP level. combination combination of complementary single marker and pathway based genome wide association analyses. This approach allowed us to identify genetic variants that carry only small disease risk individually, but that jointly can perioperative MI contribute susceptibility. Furthermore, an relatively application of results large from pathway based analysis may effects add structure to on interpreting genomic data and allow exploration of cellular processes that functionally underpin the observed associations. Finally, by using this 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from The possibility of rare genetic variants that drive a pronounced Supplemental Supplemental Results B.6. ) show that, on samplebased current
17 8 10 11 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only In In conclusion, we report the first genome wide association study in a cohort of patients at risk for MI after CABG surgery. Several limitations are worth mentioning. Power calculations ( limitations Several calculations Power worth mentioning. are implicated implicated in the regulation of extracellular matrix remodeling, as well as the ER to Golgi secretory pathway potentially involved in clinical clinical phenotype was not explored in this study, because only variants with minor allele frequencies > 0.05 were assessed. Also, drugs drugs that have already been developed. Thus, by improving risk assessment and identifying allele specific therapeutic responses, further investigation of loci and pathways prioritized in this cardioprotection. study could yield actionable results for enhancing perioperative size and incidence of perioperative MI, our study can detect a genotypic relative risk ~2 with 80% power (assuming a variant with Based Based on our integrated approach utilizing both primary (perioperative MI) and secondary phenotypes (quantitative CK MB, extreme CK MB), single marker analysis and pathway analysis, we identified several polymorphisms in the insulin growth factor system functional functional studies to further elucidate the potential biological effects of the SNPs identified were not feasible due to lack of plasma or our Caucasian, and therefore predominantly patients in our in enrolled study. cohortswere Finally, the GeneMAGIC availability tissue groups. ethnic toother be cannot generalized findings 10% 10% minor allele frequency and a realistic linkage disequilibrium between the tested marker and the causal locus D’ = 0.8). Thus, although ours is the largest genetic association study of perioperative MI conducted to date, it is powered to detect only variants with common relatively large effect sizes. Most published genetic association studies of perioperative MI have reported larger effect sizes compared to ambulatory populations (OR range 1.79 3.97). Page 17 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 18 of 52 The use of such animal models 37
18 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Currently, Currently, very few existing groups can bridge from genetics to molecular biology or cell physiology and to disease or novel therapy; we propose to examine the mechanisms by which variation in the observed genes is involved in myocardial ischemia reperfusion by injury examining their ineffects functional from patients myocardium human as cardiac undergoing wellsurgery, in our previously described preclinical rodent and swine models of cardioplegic arrest and cardiopulmonary bypass. would would allow pharmacological studies targeting the identified pathways to explore the mechanism by which novel cardioprotective drugs would attenuate myocardial ischemia reperfusion injury, with the ultimate goal of developing personalized cardioprotective adaptive adaptive responses to ER stress. As other GWAS of PMI cohorts publish their results, we intend to collaborate on conducting a meta analysis for this particular phenotype. While our GWAS results are intriguing, follow up studies are needed to translate these initial findings into the biological insights that could lead to predictive and therapeutic advances in perioperative care. For instance, in the regions of confirmed associations, causal variants will occasionally beonly among those directly genotyped. Moreover, GWAS detect almost exclusively the effects of common SNPs, offering limited power to capture any rare and structural variants, such as insertions, deletions, inversions and translocations. Detailed sequencing may be necessary to further characterize the genetic variations in the PMI associated regions to enhance the identification of causal variants. Furthermore, as in most cases, the functions of genes identified and their variants, as well as the mechanisms by which they may contribute to PMI pathophysiology, are largely unknown. strategies. strategies. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
19 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. None None For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only This work was supported, in part, by National Institutes of Health grants R01 HL075273 and R01 HL092071 (to Dr. CONTRIBUTORSHIP STATEMENT: STATEMENT: CONTRIBUTORSHIP and analysis tothe contributed YJL of manuscript. drafting the and data, of interpretation and tothe analysis contributed MDK the of data. analysis and control quality tothe and contributed YWL YL manuscript. the of revision data and critical of interpretation Podgoreanu) Podgoreanu) and by American Heart Association grants 0256342U Podgoreanu). and 9951185U (to Dr. Mathew), and 0120492U (to ETHICSAPPROVAL: Dr. Board Review Institutional School Medicine of University Duke The STATEMENT: SHARING DATA JA, MFN, PKS, DJ and JPM contributed to the acquisition of the data and critical review of the manuscript. MVP contributed to the tothe contributed MVP manuscript. the of review and critical data the acquisition of tothe JPMcontributed PKS, DJ JA,and MFN, the versionof final the approved and review manuscript critical data, the of interpretation and design, analysis and conception manuscript. submitted INTERESTS: COMPETING FUNDING: No additional data available available data additional No Page 19 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 20 of 52 (5):443 53. (5):443 53. 357 (2):483 8, 88 e1 2. e1 2. 88 (2):483 8, 139
20 BMJ Open http://bmjopen.bmj.com/ (10):1172 9. (10):1172 9. (13):1675 84. (13):1675 84. 124 117 on October 1, 2021 by guest. Protected copyright. (3):681 9. (3):681 9. (6):585 91. (6):585 91. 154 305 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (11 Suppl):S60 5. Suppl):S60 5. (11 (5830):1488 91. (5830):1488 91. (8):793 801. (8):793 801. 122 316 51 coronary artery bypass graft surgery. J Thorac Cardiovasc Surg 2010; Surg Cardiovasc J Thorac surgery. graft bypass artery coronary 2010; coronary artery disease. Circulation 2008; Circulation disease. artery coronary 2007; study. Atherosclerosis 2001; Atherosclerosis study. 2008; Blood Institute Workshop. Circulation 2011; Circulation Workshop. Institute Blood graft surgery. Jama 2011; Jama surgery. graft 8. Liu KY, Muehlschlegel JD, Perry TE, et al. Common genetic variants on chromosome 9p21 predict perioperative myocardial injury after after injury myocardial perioperative predict 9p21 chromosome on variants genetic Common al. et TE, Perry JD, Muehlschlegel KY, Liu 8. 9. Muehlschlegel JD, Liu KY, Perry TE, et al. Chromosome 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation Circulation surgery. graft bypass artery after coronary mortality predicts variant 9p21 Chromosome al. et TE, Perry KY, JD,Liu Muehlschlegel 9. 6. Echahidi N, Pibarot P, O'Hara G, et al. Mechanisms, prevention, and treatment of atrial fibrillation after cardiac surgery. J Am Coll Cardiol Coll Cardiol J Am surgery. cardiac after fibrillation atrial of treatment and prevention, Mechanisms, al. et G, O'Hara P, Pibarot N, Echahidi 6. 5. Schunkert H, Gotz A, Braund P, et al. Repeated replication and a prospective meta analysis of the association between chromosome 9p21.3 and and 9p21.3 chromosome between theassociation of meta analysis aprospective and replication Repeated al. et P, Braund A, Gotz H, Schunkert 5. 4. McPherson R, Pertsemlidis A, Kavaslar N, et al. A common allele on chromosome 9 associated with coronary heart disease. Science Science disease. heart coronary with associated 9 chromosome allele on common al.A et N, Kavaslar A, Pertsemlidis R, McPherson 4. 7. Pankow JS, Folsom AR, Cushman M, et al. Familial and genetic determinants of systemic markers of inflammation: the NHLBI family heart family NHLBI the inflammation: of markers systemic of determinants genetic and al.Familial etM, Cushman AR, Folsom JS, Pankow 7. 3. Samani NJ, Erdmann J, Hall AS, et al. Genomewide association analysis of coronary artery disease. N Engl J Med2007; Engl N disease. artery coronary of analysis association Genomewide etal. J,HallAS, Erdmann NJ, Samani 3. 2. Schwartz Longacre L, Kloner RA, Arai AE, et al. New horizons in cardioprotection: recommendations from the 2010 National Heart, Lung, and and Lung, Heart, National the 2010 from recommendations incardioprotection: horizons New al. et AE, Arai RA, Kloner L, Longacre Schwartz 2. REFERENCES: REFERENCES: bypass artery coronary following survival and elevation enzyme of myocardial Association al. V,et K, Hasselblad MJ, Mahaffey Domanski 1. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (7):776 88. (7):776 88. 31 (19):2446 54. (19):2446 54. 294 (9):1045 53. (9):1045 53. 18 (11 Suppl):I106 12. (11Suppl):I106 12. 116 (8):1072 5. (8):1072 5.
14 21 BMJ Open http://bmjopen.bmj.com/ (11 Suppl):S143 8. Suppl):S143 8. (11 124 (1 Suppl):I275 81. (1Suppl):I275 81. on October 1, 2021 by guest. Protected copyright. 114 (2):131 51. (2):131 51. 125 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (5):265 72. (5):265 72. (1):9 13. (1):9 13. 92 16 Statement. Hum Genet 2009; Genet Hum Statement. surgery in replicated analyses. Circulation 2011; Circulation analyses. in replicated surgery 2008; cardiac surgery. Circulation 2006; Circulation surgery. cardiac for assessment of the atrial fibrillation substrate. Europace 2012; Europace substrate. fibrillation atrial the of forassessment 1999; graft failure following coronary artery bypass graft surgery: PREVENT IV: a randomized controlled trial. Jama 2005; Jama trial. controlled randomized a IV: PREVENT surgery: graft bypass artery coronary following failure graft infarction in whites undergoing coronary artery bypass graft surgery. Circulation 2007; Circulation surgery. graft bypass artery coronary undergoing whites in infarction 13. Schotten U, Maesen B, Zeemering S. The need for standardization of time and frequency domain analysis of body surface electrocardiograms electrocardiograms surface of body analysis frequency domain and time of standardization for need The S. Zeemering B, Maesen U, Schotten 13. 14. Luo L, Peng G, Zhu Y, et al. Genome wide gene and pathway analysis. Eur J Hum Genet 2010; Genet Hum J Eur analysis. pathway and gene Y, Genome wide etal. Zhu G, Peng L, Luo 14. 16. Little J, Higgins JP, Ioannidis JP, et al. Strengthening the reporting of genetic association studies (STREGA): an extension of the STROBE STROBE the of extension an (STREGA): studies association of genetic the reporting al.Strengthening etJP, JP,Ioannidis J, Higgins Little 16. 12. Lobato RL, White WD, Mathew JP, et al. Thrombomodulin gene variants are associated with increased mortality after coronary artery bypass bypass artery coronary after mortality increased with associated are variants gene Thrombomodulin etJP,al. Mathew WD, White RL, Lobato 12. 15. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by genome wide association. Genomics Genomics association. genome wide by assessed diseases common seven of analysis Pathway NJ. Schork EJ, Topol A, Torkamani 15. 11. Podgoreanu MV, White WD, Morris RW, et al. Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after after infarction myocardial postoperative of risk and polymorphisms gene al.Inflammatory et RW, Morris WD, White MV, Podgoreanu 11. 17. Skol AD, Scott LJ, Abecasis GR, et al. Optimal designs for two stage genome wide association studies. Genet Epidemiol 2007; Epidemiol Genet studies. association genome wide two stage for designs Optimal al. et GR, Abecasis LJ, Scott AD, Skol 17. 19. Nashef SA, Roques F, Michel P, et al. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg Surg JCardiothorac Eur (EuroSCORE). evaluation risk operative for cardiac system European al. et P, Michel F, Roques SA, Nashef 19. 18. Alexander JH, Hafley G, Harrington RA, et al. Efficacy and safety of edifoligide, an E2F transcription factor decoy, for prevention of vein ofvein for prevention decoy, factor transcription E2F an of edifoligide, safety and Efficacy al. et RA, Harrington G, JH,Hafley Alexander 18. 10. Collard CD, Shernan SK, Fox AA, et al. The MBL2 'LYQA secretor' haplotype is an independent predictor of postoperative myocardial myocardial postoperative of predictor independent an is haplotype secretor' 'LYQA MBL2 The al. et AA, Fox SK, Shernan CD, Collard 10. Page 21 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 22 of 52 (2):229 37. (2):229 37. 45 (16):2020 35. (16):2020 35. 126
22 BMJ Open http://bmjopen.bmj.com/ :57 67. :57 67. 47 on October 1, 2021 by guest. Protected copyright. (18):e211 2; author reply e11 2. e11 2. reply author (18):e211 2; 109 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (8):904 9. (8):904 9. 38 (3):479 92. (3):479 92. (5830):1491 3. (5830):1491 3. (9654):1962 76. (9654):1962 76. (4):e61114. (4):e61114. 191 8 316 372 comparison with markers of systemic inflammation, platelet activation, and myocardial necrosis. J Am Coll Cardiol 2005; Cardiol Coll JAm necrosis. myocardial and activation, platelet inflammation, systemic of markers with comparison syndromes. Circulation 2004; Circulation syndromes. 2010; 2013; Genet 2006; Genet 2007; and human diseases. Int J Biochem Cell Biol 2014; Biol Cell J Biochem Int diseases. human and 2008; 28. Conti E, Andreotti F, Zuppi C. Pregnancy associated plasma protein a as predictor of outcome in patients with suspected acute coronary coronary acute suspected with in patients outcome of predictor aas protein plasma Pregnancy associated C. Zuppi F, Andreotti E, Conti 28. 27. Heeschen C, Dimmeler S, Hamm CW, et al. Pregnancy associated plasma protein A levels in patients with acute coronary syndromes: syndromes: coronary acute with patients inlevels protein A plasma et al. CW, Pregnancy associated Hamm S, Dimmeler C, Heeschen 27. 26. Thygesen K, Alpert JS, Jaffe AS, et al. Third Universal Definition of Myocardial Infarction. Circulation 2012; Circulation Infarction. Myocardial of Definition ThirdUniversal al. et AS, Jaffe JS, K, Alpert Thygesen 26. 23. Bangalore S, Wetterslev J, Pranesh S, et al. Perioperative beta blockers in patients having non cardiac surgery: a meta analysis. Lancet Lancet a meta analysis. surgery: non cardiac having patients in blockers beta Perioperative al. et S, J, Pranesh Wetterslev S, Bangalore 23. 22. Fox RM, Hanlon CD, Andrew DJ. The CrebA/Creb3 like transcription factors are major and direct regulators of secretory capacity. J Cell Biol Biol J Cell capacity. secretory of regulators direct and major are factors transcription CrebA/Creb3 like The DJ. Andrew CD, Hanlon RM, Fox 22. 25. Baines AC, Adams EJ, Zhang B, et al. Disruption of the Sec24d gene results in early embryonic lethality in the mouse. PLoS One One PLoS mouse. in the lethality embryonic early in results gene Sec24d ofthe et al. B, Disruption Zhang EJ, Adams AC, Baines 25. 21. Price AL, Patterson NJ, Plenge RM, et al. Principal components analysis corrects for stratification in genome wide association studies. Nat Nat studies. association ingenome wide stratification for corrects analysis components Principal al. et RM, Plenge NJ, Patterson AL, Price 21. 20. Helgadottir A, Thorleifsson G, Manolescu A, et al. A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science Science infarction. of myocardial risk the affects 9p21 chromosome on variant common A al. et A, Manolescu G, Thorleifsson A, Helgadottir 20. 24. Unlu G, Levic DS, Melville DB, et al. Trafficking mechanisms of extracellular matrix macromolecules: insights from vertebrate development development vertebrate from insights macromolecules: matrix of extracellular mechanisms Trafficking al. et DB, Melville DS, Levic G, Unlu 24. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (1):15 24. (1):15 24. 98 :6357 60. :6357 60. (4):950 6. (4):950 6. 76 2012
23 (2):85 92. (2):85 92. 1 BMJ Open http://bmjopen.bmj.com/ (9):957 70. (9):957 70. 61 (3):351 60. (3):351 60. 98 :51. :51. 3 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (2):184 95. (2):184 95. (10):3549 60. (10):3549 60. 74 22 collaborative meta analysis. J Am Coll Cardiol 2013; Coll Cardiol J Am meta analysis. collaborative methodology paper. J Cardiothorac Surg 2008; Surg Jpaper. Cardiothorac methodology chronic coronary artery disease. Circ Cardiovasc Genet 2008; Genet Cardiovasc Circ disease. artery coronary chronic infarction in patients with established coronary artery disease. Circ J 2012; Circ disease. artery coronary established with patients in infarction fibrillation wave reconstruction. Conf Proc IEEE Eng Med Biol Soc 2012; Soc MedBiol Eng IEEE Proc Conf reconstruction. wave fibrillation from ischemia/reperfusion injury. Circ Res 2006; Res Circ injury. ischemia/reperfusion from 2007; 2008; 36. Chan K, Patel RS, Newcombe P, et al. Association between the chromosome 9p21 locus and angiographic coronary artery disease burden: a burden: disease artery coronary angiographic 9p21and locus chromosome the between Association et P, al. Newcombe RS, K,Patel Chan 36. 37. de Lange F, Yoshitani K, Podgoreanu MV, et al. A novel survival model of cardioplegic arrest and cardiopulmonary bypass in rats: a in rats: bypass cardiopulmonary and arrest cardioplegic of model survival novel A al. et MV, K,Podgoreanu Yoshitani F, Lange de 37. 35. Horne BD, Carlquist JF, Muhlestein JB, et al. Association of variation in the chromosome 9p21 locus with myocardial infarction versus versus infarction myocardial with locus 9p21 chromosome the in of variation Association JB,al. et Muhlestein JF, Carlquist BD, Horne 35. 34. Virani SS, Brautbar A, Lee VV, et al. Chromosome 9p21 single nucleotide polymorphisms are not associated with recurrent myocardial myocardial recurrent with associated not are polymorphisms nucleotide single 9p21 al.Chromosome etVV, Lee A, Brautbar SS, Virani 34. 33. Zeemering S, Maesen B, Nijs J, et al. Automated quantification of atrial fibrillation complexity by probabilistic electrogram analysis and and analysis electrogram probabilistic by complexity fibrillation atrial of quantification Automated al. J,et Nijs B, Maesen S, Zeemering 33. 32. Kempf T, Eden M, Strelau J, et al. The transforming growth factor beta superfamily member growth differentiation factor 15 protects the heart the heart protects factor 15 growth differentiation member superfamily factor beta growth transforming The al. J,et Strelau M, T, Eden Kempf 32. 31. Bujak M, Frangogiannis NG. The role of TGF beta signaling in myocardial infarction and cardiac remodeling. Cardiovasc Res Res Cardiovasc remodeling. cardiac and infarction in myocardial signaling TGF beta of role The NG. Frangogiannis M, Bujak 31. 30. Backs J, Olson EN. Control of cardiac growth by histone acetylation/deacetylation. Circ Res 2006; Res Circ acetylation/deacetylation. histone by growth cardiac of Control EN. J, Olson Backs 30. 29. Granger A, Abdullah I, Huebner F, et al. Histone deacetylase inhibition reduces myocardial ischemia reperfusion injury in mice. Faseb J Faseb in mice. injury ischemia reperfusion myocardial reduces inhibition deacetylase Histone al. et F, Huebner I, Abdullah A, Granger 29. Page 23 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 24 of 52 (p) for all single nucleotide polymorphisms nucleotide single all for (p) 10 ), but 521 SNPs met the prespecified discovery threshold threshold discovery prespecified the SNPs met 521 but ), −8
, rs2044061 on chromosome 8). chromosome on rs2044061 , 6 24 <9.09×10 p BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (SNPs). None of the SNPs reached genome wide significance ( significance SNPs the genome wide of reached None (SNPs). LEGENDS FIGURE x axis The analysis. I ininfarction Stage myocardial perioperative with association of1. plot genome-wide Manhattan Figure the –log shows chromosome), by y axis order (colored in physical the genome represents p< 0.001 for inclusion in Stage II analyses (minimum p = 2.76x10 = p (minimum analyses in inclusion II Stage for 0.001 p<
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 Figure 1 22 264x105mm (72 x 72 DPI) 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 SUPPLEMENTAL MATERIAL 3 4 5 6 Table of Contents 7 8 PEGASUS Investigative Team Members Page 2 9 10 Supplemental Methods Page 3 11 12 13 A.1. Quality control of data flow Page 3 14 15 A.2. ControllingFor for populationpeer stratification review only Page 4 16 17 A.3. Haplotype analysis Page 6 18 19 A.4. GWAS pathway analysis Page 6 20 21 Supplemental Results Page 7 22 23 B.1. Patient characteristics Page 7 24 25 B.2. Single-locus GWAS analysis Page 8 26 27 B.3. Haplotype analysis Page 8 28 29 B.4. Chromosome 9p21 locus analysis Page 14 30 31 B.5. GWAS pathway analysis results Page 15 32 33
B.6. Power calculation Page 21 http://bmjopen.bmj.com/ 34 35 B.7. Details of the top 521 SNPs identified in stage I GWAS analyses Page 22 36 37 38 Supplemental References Page 25 39 40 41 42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 1 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Perioperative Genetics and Safety Outcomes Study (PEGASUS) Investigative Team Members 3 4 Allen AS, Davis RD, Funk B, Gaca JG, Ginsburg GS, Glower DD, Goldstein DB, Grichnik KP, Hall RL, 5 6 Hauser E, Jones R, Kertai MD, Laskowitz DT, Li YJ, Lodge AJ, Mathew JP, Milano CA, Moretti EW, 7 8 Newman MF, Phillips-Bute B, Podgoreanu MV, Smith MP, Smith PK, Stafford-Smith M, Swaminathan 9 10 M, Welsby IJ, White WD, Willard HF 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 2 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 A. SUPPLEMENTAL METHODS 3 4 A.1. Quality control of data flow 5 6 The following quality control steps were taken to make sure that genotypes were correctly called. 7 8 Infinium BeadStudio raw data analysis: Any sample that had a very low intensity or a very low 9 • 10 11 call rate using the standard Illumina cluster (< 95%) was deleted. All SNPs that had a call 12 13 frequency below 100% were then reclustered. Any sample that was below a 98% call rate after 14 15 reclusteringFor was deleted. peer Next, all SNPs review that had a call frequency only below 99% were deleted. Any 16 17 SNPs for which >1% of samples were not called or were ambiguously called were deleted. A 18 19 total of 35 samples (13 in GeneMagic and 22 in PEGASUS cohort) were deleted during this 20 21 procedure.! 22 23 • Minor allele frequency (MAF) check for data handling accuracy: This step performs a basic 24 25 check of the data accuracy on the data flow pipeline from the output of the Illumina genotyping 26 27 facility to the analytical process. We checked the MAF report from PLINK software against the 28 29 original locus report generated by the genotyping facility, and ensured that the 2 matched exactly.! 30 31 Gender specification: This step compares the observed genotypes on chromosomes X and Y to 32 • 33 34 the gender specification obtained from the phenotype database. Samples with X-chromosomal http://bmjopen.bmj.com/ 35 36 heterozygosity that was inconsistent with reported gender were individually inspected against the 37 38 original data source. A total of 11 samples could not be reconciled and were excluded at this step 39 40 (8 in GeneMagic and 3 in PEGASUS cohort)! 41 42 • Cryptic relatedness: We estimated the sharing of genetic information between cohort participants on October 1, 2021 by guest. Protected copyright. 43 44 by calculating identity by descent (IBD) using PLINK software. All pairs with DNA samples 45 46 showing ≥ 0.125 estimated proportion of alleles IBD were inspected, and one sample from each 47 48 pair was excluded from further analyses. A total of 9 samples (all in PEGASUS cohort) were 49 50 excluded at this step. ! 51 52 Genotype missing: This step assesses whether the genotype missing is skewed toward cases or 53 • 54 55 controls, which could give rise to spurious association p-values. We used PLINK software to 56 57 perform this check on the top SNPs discussed in the paper.! 58 59 3 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 • Low MAF: We removed all SNPs with MAF < 0.01 in order to control for error in the estimation 3 4 of asymptotic p-values, as alleles with such a low MAF would have no chance of approaching 5 6 significance based on our sample size and perioperative MI incidence.! 7 8 Hardy-Weinberg Equilibrium (HWE): We used PLINK software to identify any observed 9 •
10 -6 11 genotypes that deviated from HWE. All markers that significantly deviated from HWE (p < 10 ) 12 13 were excluded.! 14 15 Analyses were performedFor using peer PLINK 1.07 , SAS/Geneticsreview version 9.2only, and R version 2.15.1. 16 17 18 19 A.2. Controlling for population stratification 20 21 We used a modified EIGENSTRAT method to control for population stratification.1 This method derives 22 23 the principal components of the correlations among the gene variants, and corrects for those correlations 24 25 in the association tests. Population structure was investigated using the EigenSoft program. All 15 26 27 principal components (PCs) were computed for both datasets. Plots between pairwise PCs, particularly 28 29 PC1 vs PC2, were generated to determine whether any obvious outliers deviated from the main cluster, 30 31 and hence, had to be excluded from subsequent analyses (Supplemental Figure S1). 32 33 http://bmjopen.bmj.com/ 34 35 36 Supplemental Figure S1. Comparison of PC1 37 38 vs PC2 reveals potential population 39 40 41 stratification in the discovery (Gene-Magic)
42 on October 1, 2021 by guest. Protected copyright. 43 dataset. 44 45 46 47 48 49 50
51 52 53
54 55 56 57 58 59 4 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 In addition, we performed multivariate regression analyses using a strategy to keep all PCs up to 6 7 the last PC with p < 0.05 in each step, to determine the number of PCs to be used for correction in the 8 9 final analysis. For instance, starting with 15 PCs in the model, if PC(i) is the last PC with p < 0.05, we 10 11 included PC(1) to PC(i) in the next multivariate model, and then repeated the process until the last PC 12 13 remained nominally significant. Using this iterative multivariate analysis, we found that the top 10 PCs 14 15 were a reasonableFor set for the peer Gene-Magic dataset review (p = 0.032), and wereonly subsequently re-estimated and 16 17 18 used for stage II analysis (Supplemental Table S1). This final set of 10 PCs was then used as covariates 19 20 to adjust for ancestry, along with clinical variables, in multivariate logistic regression analysis of 21 22 perioperative MI. 23 24 25 Supplemental Table S1: Principal component analysis of the study populations 26 Supplemental Table S1. Principal Component Analysis of the Study Populations
27 ANALYSIS OF MAXIMUM LIKELIHOOD ESTIMATES 28 29 DISCOVERY DATSET REPLICATION DATASET 30 Parameter DF Estimate Standard Wald chi- P-value DF Estimate Standard Wald chi- P-value 31 error square test error square test 32 33 Intercept 1 -2.635 0.127 434.325 <0.0001 1 -2.153 0.076 798.277 <0.0001 http://bmjopen.bmj.com/ 34 Principal component 1 1 -31.474 14.214 4.903 0.027 1 18.429 8.413 4.798 0.029 35 Principal component 2 1 -5.486 3.826 2.056 0.152 1 -4.009 3.338 1.443 0.230 36 37 Principal component 3 1 -7.575 3.497 4.691 0.03 1 -3.239 3.230 1.006 0.316 38 Principal component 4 1 -0.431 3.479 0.015 0.901 1 -1.266 3.119 0.165 0.685 39 40 Principal component 5 1 18.941 9.623 3.874 0.049 1 -1.642 4.828 0.116 0.734 41 Principal component 6 1 0.519 7.944 0.004 0.948 1 -15.494 7.678 4.073 0.044 on October 1, 2021 by guest. Protected copyright. 42 Principal component 7 1 -2.231 10.001 0.049 0.824 1 12.985 6.873 3.569 0.059 43 44 Principal component 8 1 -10.793 9.043 1.424 0.233 1 2.244 4.825 0.216 0.642 45 Principal component 9 1 29.876 11.661 6.564 0.01 1 -8.120 5.366 2.289 0.130
46 Principal component 10 1 -22.775 10.593 4.623 0.032 1 7.487 4.650 2.592 0.107 47 48 DF, degrees of freedom 49 50 51 16 52 53 54 55 56 57 58 59 5 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 A.3. Haplotype analysis 5 6 For the final candidate genes prioritized based on the significance of SNP association tests, we first 7 8 identified the linkage disequilibrium (LD) blocks harboring the significant SNPs. Within each LD block, 9 10 we performed haplotype association analysis using sliding windows of 3 markers across the region. This 11 12 was conducted for the expanded dataset, by constructing multivariable logistic regression models 13 14 adjusting for the same clinical covariates and PCs used for single marker analyses. 15 For peer review only 16 17 18 19 A.4. GWAS pathway analysis 20 21 SNPs with a covariate-adjusted association p-value < 0.01 in stage I GWAS were subjected to functional 22 23 ontology enrichment analyses using tools implemented in the MetaCore software suite (GeneGO, St. 24 25 Joseph, MI). Enrichment analysis begins by assigning SNPs to genes, and matching these geneIDs with 26 27 geneIDs in functional ontologies in highly curated canonical pathway maps in MetaCore. The aggregation 28 29 of GWAS associations for the genes in tested pathways, ie, the significance of enrichment, is computed as 30 31 the p-value of a hypergeometric model.2 Canonical pathways enriched in stage I and stage II analyses 32 33
were further compared; and the probability of a random intersection between the set of genes in common http://bmjopen.bmj.com/ 34 35 in the 2 cohorts, and functional ontology entities was estimated as p-value of a hypergeometric 36 37 distribution in MetaCore. 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 6 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B. SUPPLEMENTAL RESULTS 3 4 B.1. Patient characteristics 5 6 Supplemental Table S2. Clinical characteristics of the study populations 7 Table 1. Clinical Characteristics of the Study Populations
8 STAGE I ANALYSIS (N=1433) STAGE II ANALYSIS (N=2055) 9 No PMI, PMI No PMI PMI 10 Characteristics OR (95%CI) P-value* OR (95%CI) P-value* (n=1321) (n=112) (n=1830) (n=225) 11 12 Extracardiac arteriopathy 251 (19.0) 20 (17.9) 0.93 (0.56-1.53) 0.77 754 (41.2) 128 (56.9) 1.88 (1.42-2.49) <0.0001 13 Recent myocardial infarction 256 (19.4) 23 (20.5) 1.08 (0.67-1.74) 0.77 393 (21.5) 39 (17.3) 0.77 (0.53-1.10) 0.15 14 Procedure other than CABG only 126 (9.5) 23 (20.5) 2.45 (1.50-4.02) 0.0004 153 (8.4) 51 (22.7) 3.21 (2.26-4.58) <0.0001 15 Year of surgery by yearFor peer review only 16 1997 26 (1.4) 3 (1.3) 1.0 17 1998 70 (3.8) 8 (3.6) 0.99 (0.24-4.02) 0.98 18 1999 87 (4.8) 3 (1.3) 0.30 (0.06-1.57) 0.15 19 2000 77 (4.2) 14 (6.2) 1.58 (0.42-5.92) 0.50 20 2001 49 (2.7) 12 (5.3) 2.12 (0.55-8.20) 0.28 21 2002 146 (11.1) 9 (8.0) 1.0 188 (10.3) 20 (8.9) 0.92 (0.26-3.32) 0.90 22 2003 1175 (88.9) 103 (92.0) 1.42 (0.70-2.87) 0.33 1189 (65.0) 109 (48.4) 0.80 (0.24-2.67) 0.71 23 2004 32 (1.8) 8 (3.6) 2.17 (0.52-9.0) 0.29 24 2005 80 (4.4) 36 (16.0) 3.90 (1.11-13.72) 0.03 25 2006 32 (1.8) 12 (5.3) 3.25 (0.83-12.75) 0.09 26 Number of diseased vessels 27 0 67 (5.1) 6 (5.3) 1.0 69 (3.8) 6 (2.7) 1.0 28 1 259 (19.6) 29 (25.9) 1.25 (0.50-3.14) 0.63 276 (15.0) 32 (14.2) 1.33 (0.54-3.32) 0.54 29 2 519 (39.3) 43 (38.4) 0.93 (0.38-2.26) 0.86 595 (32.5) 52 (23.1) 1.01 (0.42-2.43) 0.99 30 3 476 (36.0) 34 (30.4) 0.80 (0.32-1.97) 0.62 885 (48.4) 130 (57.8) 1.69 (0.72-3.97) 0.22 31 4 5 (0.3) 5 (2.2) 11.5 (2.58-51.24) 0.0014 32 Aortic cross clamp time, min 51.2+36.8 71.9+32.4 1.015 (1.01-1.02) <0.0001 55.0+35.2 80.5+36.7 1.019 (1.015-1.023) <0.0001 33
Continuous variables are presented as means+standard deviation, and categorical variables as percent frequencies. CABG, coronary artery bypass http://bmjopen.bmj.com/ 34 35 Descriptivesurgery; OR (95%CI), statistics univariate of odds clinical ratio (95% variables confidence are interval) presented; *Comparisons as countmade using (percent the t test or frequency) chi-square test, for as appropriate categorical 36 37 variables and mean ± SD for continuous variables. PMI, postoperative myocardial infarction; CABG, 38 39 coronary artery bypass surgery; extracardiac arteriopathy, defined as any one or more of the following: 40 41 claudication, carotid occlusion or > 50% stenosis, and/or previous or planned intervention on the 42 on October 1, 2021 by guest. Protected copyright. 43 abdominal aorta, limb arteries or carotids; OR (95%CI), univariate odds ratio (95% confidence interval). 44 45 *Comparisons made using the t-test or Wald chi-square test, as appropriate. Clinical variables were 46 47 included in the final multivariate logistic regression analysis if significant (p < 0.05) in univariate 48 49 analyses. 50 51
52 53 54 55 56 57 58 59 7 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.2. Single-locus GWAS analysis - population structure adjustment 3 4 5 6 7 8 Supplemental Figure S2. Quantile - Quantile (Q-Q) 9 10 plot representation of the GWAS association results 11 12 13 in the discovery dataset 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 Distribution of expected (under the null hypothesis) vs observed p-values for single nucleotide 26 27 polymorphisms that are associated with perioperative myocardial infarction, based on a whole genome- 28 29 wide analysis. The red diagonal line represents the line obtained if the observed distribution did not 30 31 deviate from the expected distribution. All p-values were corrected for the inflation factor (λ = 1.0073). 32 33
http://bmjopen.bmj.com/ 34 35 B.3. Haplotype analysis of genes harboring top scoring SNPs 36 37 38 Supplemental Figure S3. Linkage disequilibrium (LD) structures of [A] PAPPA2 (1q23-q25 39 40 chromosome region), [B] SEC24D (4q26 chromosome region), and [C] HDAC4 (2q37.3 chromosome 41
42 region), and displayed as pairwise correlation plots using HapMap database (CEU European ancestry) on October 1, 2021 by guest. Protected copyright. 43 44 (Haploview). Regions of LD are shaded in bright red (strong LD) and lighter for moderate or weak LD. 45 46 The physical location of the individual SNPs associated with perioperative myocardial infarction in 2- 47 48 stage analyses is circled in red. True haplotype blocks in the population are marked with black lines in the 49 50 correlation plot. 51 52 53 54 55 56 57 58 59 8 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 A. PAPPA2 gene 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 B. SEC24D gene 23 24
25 26
27 28
29 30
31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 C. HDAC4 gene 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 9 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 05 06 06 05 06 - - - - -
10 P 0.589 0.282 0.141 0.123 0.266 0.266 0.917 0.936 0.0182 0.0184 0.0345 0.0754 0.0754 0.00199 0.00198 0.00075 1.01E 4.98E 4.98E 1.45E 6.47E 11 0.000477 0.000477 0.000248 0.000256 0.000274 0.000796 12
13 14 19.5 9.56 9.56 5.58 1.16 20.8 5.55 2.17 20.8 4.47 3.16 12.2 3.16 2.38 18.8 12.2 1.24 20.3 13.4 1.24 13.4 11.4 13.2 11.3 0.292 0.0109 STAT 15 For peer review only 0.00645 16
17
18 0.5 OR 1.43 1.08 1.29 1.22 1.29 1.19 1.25 1.23 1.23 1.17 1.58 1.28 1.62 1.28 1.49 1.02 1.49 1.02 19 0.464 0.698 0.474 0.474 0.631 0.515 0.616 0.618 20
21
22 F 0.166 0.655 0.179 0.344 0.576 0.175 0.578 0.248 0.175 0.589 0.237 0.241 0.237 0.522 0.193 0.759 0.197 0.755 0.245 0.675 0.244 0.676
23 0.0786 0.0485 0.0485 0.0795 0.0801 24
25 26
27 AA AA AA AC CA AC CC GA AG GA GA AG AG GA GA GA AG GA AG GC CG GG GG GG GG GG GG
28 29
30 HAPLOTYPE 31 10
32 33 34 http://bmjopen.bmj.com/ SNP2
35 RS909941 RS909941 RS909941 RS240104 RS240104 RS240104 RS1395460 RS1395460 RS1395460 RS3795320 RS3795320 RS3795320 RS7529392 RS7529392 RS7529392 36 RS10489478 RS10489478 RS10489478 RS10798463 RS10798463 RS10798463 RS12139318 RS12139318 RS12139318 RS12565327 RS12565327 RS12565327
PAPPA2 and HDAC4 and PAPPA2
37 38 39 SNP1
40 RS909941 RS909941 RS909941 RS240104 RS240104 RS240104 RS1395460 RS1395460 RS1395460 RS3795320 RS3795320 RS3795320 41 RS10489477 RS10489477 RS10489477 RS10489478 RS10489478 RS10489478 RS10798463 RS10798463 RS10798463 RS12139318 RS12139318 RS12139318 RS12565327 RS12565327 RS12565327
42 on October 1, 2021 by guest. Protected copyright.
43
44 BP2
45 176562796 176562796 176562796 176565307 176565307 176565307 176566350 176566350 176566350 176571431 176571431 176571431 176571578 176571578 176571578 176589732 176589732 176589732 176591291 176591291 176591291 176593524 176593524 176593524 176603840 176603840 176603840 46 47
48 BP1
49 176562608 176562608 176562608 176562796 176562796 176562796 176565307 176565307 176565307 176566350 176566350 176566350 176571431 176571431 176571431 176571578 176571578 176571578 176589732 176589732 176589732 176591291 176591291 176591291 176593524 176593524 176593524 50
Results of the haplotype analyses for for analyses haplotype the of Results
51 . 3
52 53 54 Table S Table PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 55 56 57 GENESYMBOL " " " " " " " " " " " " " " " " " " " " " " " " " " "
58 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 59
60 CHR Supplemental For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 06 06 06 06 06 05 05 ------10 0.288 0.288 0.174 0.239 0.239 0.231 0.119 0.234 0.362 0.464 0.749 0.0145 0.0145 0.0238 0.0787 0.0238 0.0146 0.0806 0.0503 0.00252 2.52E 2.56E 2.65E 2.60E 2.59E 1.79E 1.58E 11 0.000264 0.000269 0.000484 0.000371 12
13 14 22.2 13.3 1.13 22.1 13.3 1.13 22.1 5.98 1.85 12.2 5.98 1.39 22.1 9.13 1.39 22.1 5.11 3.09 12.7 5.11 1.44 2.43 1.42 5.97 3.05 18.4 3.83 18.6 15 For peer review only0.831 0.537 0.102 16
17
18 1.2 0.46 1.62 1.23 0.46 1.62 1.23 1.29 1.16 1.29 1.17 1.42 1.17 1.28 1.23 1.28 1.18 1.18 1.23 1.17 2.18 2.19 19 0.461 0.665 0.461 0.461 0.625 0.832 0.721 0.804 0.962 20
21 22 0.177 0.756 0.176 0.756 0.176 0.489 0.336 0.351 0.489 0.161 0.176 0.663 0.161 0.176 0.591 0.232 0.245 0.591 0.164 0.769 0.164 0.227 0.706 0.061 0.663 0.276 0.061
23 0.0675 0.0674 0.0665 0.0615 24 25 26
27 AA AA AA AA AA AA AA AA CA GA AG AG GA GA GA AG GA AG AG GA AA GC AG AG 28 GG GG GG GG GG GG GG 29 30 31 11
32 33 34 http://bmjopen.bmj.com/
35 RS726252 RS726252 RS726252 RS791031 RS791031 RS791031 RS4507975 RS4507975 RS4507975 RS2048765 RS2048765 RS2048765 RS6737742 36 RS12133739 RS12133739 RS12133739 RS10913237 RS10913237 RS10913237 RS10798473 RS10798473 RS10798473 RS17546000 RS17546000 RS17546000 RS10454444 RS10454444 RS10454444 RS12138021 RS12138021 RS12138021
37 38 39
40 RS726252 RS726252 RS726252 RS791031 RS791031 RS791031 RS7529392 RS7529392 RS7529392 RS2048765 41 RS12133739 RS12133739 RS12133739 RS10913237 RS10913237 RS10913237 RS10798473 RS10798473 RS10798473 RS17546000 RS17546000 RS17546000 RS10454444 RS10454444 RS10454444 RS12138021 RS12138021 RS12138021 RS13414769 RS13414769 RS13414769
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 176604564 176604564 176604564 176630416 176630416 176630416 176646383 176646383 176646383 176647000 176647000 176647000 176649181 176649181 176649181 176660247 176660247 176660247 176667810 176667810 176667810 176680137 176680137 176680137 176684432 176684432 176684432 240208154 240208154 240208154 240216021 46 47 48
49 176603840 176603840 176603840 176604564 176604564 176604564 176630416 176630416 176630416 176646383 176646383 176646383 176647000 176647000 176647000 176649181 176649181 176649181 176660247 176660247 176660247 176667810 176667810 176667810 176680137 176680137 176680137 240201797 240201797 240201797 240208154 50 51
52
53 54 HDAC4 HDAC4 HDAC4 HDAC4 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 55 PAPPA2 PAPPA2 PAPPA2 PAPPA2 56 57
58 " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 06 05 05 - - - 10 0.91 0.96 0.95 0.913 0.956 0.961 0.986 0.015 0.032 0.862 0.011 0.867 0.758 0.737 0.108 0.593 0.533 0.583 0.895 0.365 0.534 0.895 0.364 0.487 0.0244 0.0575 0.0689 9.58E 2.32E 2.43E 11 0.000026 12
13
14 4.6 5.06 5.92 19.6 6.47 17.7 3.61 17.9 3.31 17.8 2.58 0.012 0.112 0.286 0.388 0.301 0.819 0.387 0.823 0.482 0.0128 0.0024 0.0301 0.0282 0.0946 0.0174 0.0174 15 0.00304 0.00252 For0.00393 0.00029 peer review only 16
17
18 1 0.8 1.01 1.01 1.01 1.32 2.21 0.75 1.02 2.14 2.13 2.12 0.84 1.06 1.07 1.09 19 0.773 0.986 0.986 0.995 0.978 0.812 0.968 0.819 0.965 0.945 0.905 0.986 0.827 0.986 0.827 20
21 22 0.51 0.73 0.52 0.13 0.718 0.221 0.221 0.185 0.594 0.182 0.594 0.224 0.266 0.221 0.208 0.414 0.524 0.415 0.423 0.512 0.488 0.382 0.546 0.382 0.723
23 0.0614 0.0615 0.0645 0.0649 0.0721 0.0721 24 25 26
27 AA AA AA AC CA AG AG GA GA AG AG GA GA AG GA AG GA GA AG AG GC CG GA 28 GG GG GG GG GG GG GG GG 29 30 31 12
32 33 34 http://bmjopen.bmj.com/
35 RS908262 RS908262 RS908262 RS925738 RS925738 RS925738 RS908265 RS908265 RS908265 RS908263 RS908263 RS908263 RS6737742 RS6737742 RS6543523 RS6543523 RS6543523 RS7594398 RS7594398 RS7594398 RS1399629 RS1399629 RS1399629 36 RS11883623 RS11883623 RS11883623 RS10200850 RS10200850 RS10200850 RS10211084 RS10211084
37 38 39
40 RS908262 RS908262 RS908262 RS925738 RS925738 RS925738 RS908265 RS908265 RS908265 RS908263 RS908263 RS2048765 RS2048765 RS6737742 RS6737742 RS6737742 RS6543523 RS6543523 RS6543523 RS7594398 RS7594398 RS7594398 RS1399629 RS1399629 RS1399629 41 RS11883623 RS11883623 RS11883623 RS10200850 RS10200850 RS10200850
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 240216021 240216021 240217116 240217116 240217116 240217522 240217522 240217522 240228263 240228263 240228263 240234901 240234901 240234901 240241255 240241255 240241255 240250456 240250456 240250456 240257958 240257958 240257958 240262287 240262287 240262287 240263103 240263103 240263103 240263584 240263584 46 47 48
49 240208154 240208154 240216021 240216021 240216021 240217116 240217116 240217116 240217522 240217522 240217522 240228263 240228263 240228263 240234901 240234901 240234901 240241255 240241255 240241255 240250456 240250456 240250456 240257958 240257958 240257958 240262287 240262287 240262287 240263103 240263103 50 51 52
53 54 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 55 HDAC4 56 57
58 " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 10 0.862 0.862 0.875 11 0.0118 0.0129 0.0122 0.0238 12
13 14 6.34 6.18 6.28 5.11 15 0.0302 0.0302 For0.0248 peer review only 16
17
18 1.29 1.27 1.02 19 0.978 0.747 0.978 0.748 20
21 22 0.31 0.52 0.17 23 0.205 0.485 0.205 0.309 24 25 26
27 AA GA GA AG GA 28 GG GG 29 30 31 13
32 33 34 http://bmjopen.bmj.com/ 35 RS6706275 RS6706275 RS6706275 36 RS10211084 RS12988669 RS12988669 RS12988669
37 38 39
40 RS908263 RS6706275 RS6706275 RS6706275 41 RS10211084 RS10211084 RS10211084
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 240263584 240265617 240265617 240265617 240275570 240275570 240275570 46 47 48
49 240263103 240263584 240263584 240263584 240265617 240265617 240265617 50 51 52
53 54 55 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 56 57
58 " " " " " " " 2 2 2 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.4. Chromosome 9p21 locus and perioperative MI 3 4 Supplemental Figure S4. Single nucleotide polymorphisms at 9p21 locus previously associated with 5 6 incident myocardial infarction3 and mortality4 after CABG. 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 The dashed horizontal line indicates the nominal significance threshold, p = 0.05. The green dots 37 38 represent single nucleotide polymorphisms (base pair) at the 9p21 locus previously identified as 39 40 predictors of perioperative MI: rs1547705 (22082375, intron); rs1333040 (22083404, intron); rs4977574 41 on October 1, 2021 by guest. Protected copyright. 42 (22098574, intron); rs944797 (22115286, intron); rs2383207 (22115959, intron); rs1537375 (22116071, 43 44 intron). The red dot represents rs10116277 (22081397) variant, which was previously associated with 45 46 47 myocardial infarction in both non-surgical and cardiac surgical cohorts. The dark blue point rs1381923 48 49 (22909438, unknown function) is the most significant variant in this plotted region. All p-values are the 50 51 results of multivariate risk factor-adjusted analyses. 52 53 54 55 56 57 58 59 14 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.5. GWAS pathway analysis results 3 4 Of the 534,350 SNPs included in analysis, 250,740 were recognized, and 66,893 were mapped to 19,163 5 6 genes within MetaCore. Stage I cohort analysis in MetaCore identified several canonical pathways after 7 8 functional enrichment analysis with SNPs that passed the p-value and odds-ratio filters (p < 0.01, 0.9 < 9 10 OR > 1.1). 11 12 13 Supplemental Figure S5. GenGo Canonical Pathway Maps histograms after enrichment analysis with 14 15 SNPs that passedFor the threshold peer and p-value filter review in GWAS stage I analyses only. 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
http://bmjopen.bmj.com/ 34 35
36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 The top 10 canonical maps for enriched SNPs are shown as histogram plots. Results are tabulated/sorted 49 50 in the histogram based on calculated –log(p-values) of hypergeometric distribution statistics. 51 52 53 54 55 56 57 58 59 15 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Figure S6. The top scoring pathway map based on the enrichment distribution in GWAS 3 4 stage I analysis – Cell adhesion: extracellular matrix remodeling. 5 6 7 8 9 10
11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
http://bmjopen.bmj.com/ 34 35
36 37 38 39 40 41
42 Top-scoring SNPs based on GWAS stage I results are linked to genes and visualized on the maps as on October 1, 2021 by guest. Protected copyright. 43 44 thermometer-like symbols. Up-ward thermometers are red and indicate an odds ratio >1 for association 45 46 with perioperative MI, and down-ward (blue) ones indicate an odds ratio for association < 1. Letters “B”, 47 48 and “C” indicate physical interactions, B, binding; and C, cleavage. Arrows with hexagons indicate 49 50 positive (green), negative (red), and unspecified (gray) effects. Arrows without hexagons indicate 51 52 technical link (gray). For further information on pathway elements please see: 53 54 http://pathwaymaps.com/pdf/MC_legend.pdf 55 56 57 58 59 16 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Top-scoring pathways in stage I analyses were compared with enriched pathways in the stage II cohort. 3 4 The probability of a random intersection between a set of genes common to both cohorts with functional 5 6 ontology entities was estimated as p-value of hypergeometric distribution using the Compare Experiments 7 8 Workflow in MetaCore. The top 4 scored canonical pathway maps (lowest p-values), based on the 9 10 enrichment distribution, are presented below. 11 12 -4 13 1. Immune response_NFAT signaling and leukocyte interaction (p = 7.2*10 ). Nuclear factors of 14 15 activated T-cellsFor (NFATs) ispeer a family of calcium review-dependent transcription only factors with pivotal roles in 16 17 regulating immune responses, which include interactions between antigen-presenting cells and other 18 19 leukocytes, and expression of a variety of cytokines by coupling changes in intracellular calcium 20 21 concentration to gene expression. Variants in CD247, calcium channels, MHC class II, and NFAT that 22 23 map to this pathway, and their association with perioperative MI is presented below in Supplemental 24 25 Table S4. 26 27 28 SNPid Gene Symbol Gene Name OR p-value 29 30 rs2982480;rs2984800;rs2995054 CD247 CD247 molecule 2.16 0.00311 31 rs42051;rs7804449;rs1544462;rs1156325; calcium channel, voltage-dependent, alpha 2/delta 32 CACNA2D1 0.71 0.00368 rs2237526;rs2018982 subunit 1 33 34 rs1779244;rs2068357 CACNB2 calcium channel, voltage-dependent, beta 2 subunit 0.50 0.00162 http://bmjopen.bmj.com/ 35 major histocompatibility complex, class II, DP beta 2 rs3117016;rs9380342 HLA-DPB2 2.45 0.00795 36 (pseudogene) 37 rs3129882 HLA-DRA major histocompatibility complex, class II, DR alpha 1.49 0.00580 38 nuclear factor of activated T-cells, cytoplasmic, rs1562724;rs1017860;rs8090864 NFATC1 1.45 0.00362 39 calcineurin-dependent 1 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 2. Cell adhesion and extracellular matrix (ECM) remodeling (p = 1.1*10-3). An important pathway 44 45 involved in responses to myocardial injury, is comprised of matrix metalloproteinases (MMPs), their 46 47 endogenous tissue inhibitors (TIMPs), and various collagens. Furthermore, the plasminogen-plasmin 48 49 system and its regulators (PAI-1) are implicated in proteolytic degradation of ECM. Finally, 2 categories 50 51 of growth factors are part of this canonical pathway – insulin-like growth factors 1 and 2 (IGF-1 and IGF- 52 53 2), and heparin-binding EGF-like growth factor (HB-EGF) – and all are involved in regulating cell 54 55 growth, proliferation, and survival. IGF-1 and IGF-2 function by activating the IGF-1 receptor, but their 56 57 58 59 17 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 bioavailability is regulated by a family of insulin-like growth factor binding proteins (IGFBP4)., HB-EGF 3 4 activates the epidermal growth factor receptor (EGFR) and v-erb-a erythroblastic leukemia viral oncogene 5 6 homolog 4 (ErbB4), leading to cell proliferation, cell survival, and tissue remodeling (Supplemental 7 8 Table S5). 9 10
11 12 SNPid Gene Symbol Gene Name OR P-value 13 14 rs953386;rs2391823 COL4A1 collagen, type IV, alpha 1 1.76 0.00162 epidermal growth factor 15 rs884904;rs845552For peerEGFR review only1.67 0.00238 receptor 16 rs9288433;rs3817429;rs2371276;rs 17 1473636;rs7425448;rs10932384;rs6 v-erb-a erythroblastic 18 740117;rs12053361;rs17803152;rs7 ERBB4 leukemia viral oncogene 0.53 0.00613 594456;rs4672615;rs1851169;rs674 homolog 4 (avian) 19 7637 20 insulin-like growth factor 1 rs2033178 IGF1 0.50 0.00901 21 (somatomedin C) 22 rs8038056;rs8028620;rs2139924;rs insulin-like growth factor 1 IGF1R 1.90 0.00383 23 3784604 receptor 24 insulin-like growth factor 2 25 rs734351 IGF2 1.57 0.00159 26 (somatomedin A) 27 rs12454479;rs8086477;rs12454564 LAMA3 laminin, alpha 3 2.01 0.00165 28 matrix metallopeptidase 29 rs762052 MMP14 1.60 0.00863 14 (membrane-inserted) 30 31 serpin peptidase inhibitor, clade E (nexin, 32 rs1050813 SERPINE1 plasminogen activator 1.38 0.00874 33 inhibitor type 1), member
http://bmjopen.bmj.com/ 34 1 35 rs783147 PLG plasminogen 0.68 0.00979
36 37 38 3. Regulation of lipid metabolism_RXR-dependent regulation of lipid metabolism via PPAR, RAR, and
39 -3 40 VDR (p = 5.1*10 ; Supplemental Table S6). 41
42 on October 1, 2021 by guest. Protected copyright. 43 SNPid Gene Symbol Gene Name OR P-value 44 carnitine rs12364396 CPT1A 1.779 0.006433 45 palmitoyltransferase 1A 46 47 rs1286767;rs216 retinoic acid receptor, RARB 1.647 0.005116 48 4360;rs2033447 beta 49 50
51 -3 52 4. BMP (bone morphogenic protein) signaling pathway (p = 5.2*10 ). BMPs are members of the TGF-β 53 54 superfamily, with important roles in embryonic development including cardiomyogenesis and apoptosis. 55 56 BMPs transduce their signals via 2 pathways: SMAD-dependent (including SMAD9[8]), and SMAD- 57 58 59 18 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 independent cascade. BMP activity can be regulated by intracellular inhibitory SMADs (like SMAD6) or 3 4 extracellularly by secreted antagonists (like gremlin). Variants in SMAD9 and SMAD6 were associated 5 6 with an increased incidence of perioperative MI (Supplemental Table S7). 7 8 9 SNPid Gene Symbol Gene Name OR p-value 10 11 12 rs11858577 SMAD6 SMAD, mothers against DPP homolog 6 2.30 0.0002506 13 Mothers against decapentaplegic, drosophila, rs7998663 SMAD9 2.47 0.000167 14 homolog of, 9 15 rs884940For SKIpeerski oncoprotein review only0.61 0.0006749 16 17 18 19 In summary, functional analysis of genome-wide association data using biological ontologies allowed us 20 21 to supplement our single marker analysis results by identifying a number of additional candidates for 22 23 further investigation as biomarkers for perioperative MI. It also highlighted the potential mechanistic 24 25 importance of several signaling pathways that have not previously been implicated in the complex and 26 27 multifactorial pathogenesis of perioperative ischemia-reperfusion injury. 28 29 30 31 32 Supplemental Figure S7 summarizes RegulomeDB (regulome.stanford.edu) and HaploReg v2 33 34 (http://www.broadinstitute.org/mammals/haploreg/haploreg.php) database results for rs609408 (intergenic http://bmjopen.bmj.com/ 35 36 variant, located 1.5kb 3’ of SMAD9). 37 38 A. RegulomeDB supporting data for rs609418 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50
51 52
53 54 55 56 57 58 59 19 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B. HaploReg v2 summary for rs609418 and variants with r2≥0.8 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 and detailed view for rs609418 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46
47 48 49 50 51 52 53 54 55 56 57 58 59 20 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.6. Power calculation 3 4 We used Genetic Power Calculator (http://pngu.mgh.harvard.edu/~purcell/gpc/) to estimate the effect size 5 6 that our study sample size can detect with 80% power. As an example, we used rs4834703 in the SEC24D 7 8 gene, which has 0.1 MAF and odds ratio (OR) of 1.98 from the stage II analysis. In this power calculation, 9 10 we assumed 11% disease prevalence, 0.1 MAF for the marker, and the same study sample sizes (225 11 12 13 cases and 1830 controls). We varied the minor allele frequency of the disease locus (0.05, 0.1, 0.15) to 14 15 assess the genotypicFor relative riskpeer (GRR) that ourreview study sample size canonly detect with 80% power. We used 16 -5 17 the same significant threshold described in the paper, 9 x 10 , for the stage II analysis. We assumed an 18 19 additive model to define GRR of Aa and AA genotypes, that is, GRR_Aa = x and GRR_Aa = 2x, where x 20 21 is equivalent to the OR for the minor allele. If the marker tested was in complete linkage disequilibrium 22 23 (D’ = 1) with the disease locus, we found that our sample size has approximately 80% power to detect a 24 25 GRR = 2.6 for a rare disease locus (MAF = 0.05) and GRR = 1.8 for a more common disease locus (MAF 26 27 = 0.1 and 0.15). The effect sizes become slightly higher when D’ is assumed at 0.8 between disease locus 28 29 and marker (Supplemental Table S8). The OR for rs4834703 in the SEC24D gene in our study was 1.97, 30 31 which is consistent with the estimate here. Although somewhat controversial, current evidence supports 32 33
joint 2-stage analysis designs over replication analyses in GWAS, based on increased statistical power of http://bmjopen.bmj.com/ 34 35 joint analyses,5, 6 which formed the rationale for our study design. 36 37 38 Supplemental Table S8. GRR estimates that our dataset can detect a given disease locus MAF at 39 40 approximately 80% power when marker MAF = 0.1, sample size = 225 cases and 1830 controls, and 41
42 disease prevalence = 11%. on October 1, 2021 by guest. Protected copyright. 43 44 D' Disease locus MAF Odds Ratio Power 45 1 0.05 2.6 0.8 46 47 0.1 1.8 0.84 48 0.15 1.75 0.81 49 0.8 0.05 3.1 0.8 50 0.1 2.15 0.84 51 0.15 2 0.82 52 53 54 55 56 57 58 59 21 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.7. Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 8 RS2044061 19723503 INTS10 | LPL G 0.3117 2.004 1.499 2.68 2.76E-06 4 RS6839725 21870314 LOC100131836 | KCNIP4 A 0.09312 2.101 1.42 3.11 0.0002053 1 RS6660197 200703541 DDX59 | CAMSAP1L1 A 0.1266 2.368 1.63 3.438 5.99E-06 11 RS1816775 13783230 FAR1 | SPON1 A 0.3728 1.71 1.288 2.269 0.0002075 1 RS12756886 200840467 CAMSAP1L1 | GPR25 G 0.1215 2.342 1.612 3.404 8.12E-06 4 RS10516398 21746541 KCNIP4 A 0.09791 2.085 1.414 3.074 0.0002086 7 3 RS6788848 103028245 LOC100287880 | ALCAM G 0.237 1.995 1.473 2.702 8.15E-06 15 RS4288951 27751567 GABRG3 A 0.08947 2.283 1.476 3.532 0.0002089 1 RS12060387 210819848 HHAT A 0.05196 3.218 1.925 5.379 8.27E-06 3 RS9883142 7017919 GRM7 A 0.3052 1.717 1.29 2.285 0.0002102 8 1 RS12742404 200811765 CAMSAP1L1 G 0.1204 2.338 1.609 3.397 8.44E-06 3 RS12631018 116501764 LOC285194 | IGSF11 G 0.1684 1.913 1.358 2.696 0.0002105 6 RS3861449 148300251 SAMD5 | SASH1 A 0.2655 1.985 1.467 2.686 8.77E-06 7 RS17171847 41370672 C7orf10 | INHBA G 0.09312 2.121 1.425 3.157 0.0002106 9 15 RS894157 90168108 C15orf42 G 0.1012 2.333 1.604 3.394 9.42E-06 15 RS6494826 70049111 LOC145837 | C15orf50 A 0.4318 1.738 1.297 2.33 0.0002164 8 RS2235118 2027154 MYOM2 C 0.1319 2.184 1.536 3.107 1.38E-05 10 RS984668 124120559 BTBD16 | PLEKHA1 A 0.3542 1.736 1.296 2.325 0.0002167 10 1 RS2292096 200826769 CAMSAP1L1 G 0.1279 2.282 1.569 3.318 1.57E-05 17 RS10512598 73038132 ATP5H A 0.05634 2.532 1.547 4.144 0.0002202 4 RS4689485 6578487 MAN2B2 A 0.1606 2.054 1.479 2.851 1.71E-05 2 RS7594398 240250456 HDAC4 A 0.0695 2.28 1.47 3.536 0.000232 11 5 RS7721080 31973116 PDZD2 G 0.04521 3.173 1.868 5.391 1.94E-05 1 RS12032817 83177365 LPHN2 | TTLL7 G 0.3934 1.703 1.283 2.261 0.0002321 6 RS9382274 53877181 LRRC1 | C6orf142 A 0.07399 2.601 1.673 4.044 2.20E-05 5 RS17105323 146114766 PPP2R2B G 0.03205 3.205 1.723 5.96 0.0002336 12 3 RS1038517 59949140 FHIT A 0.0898 2.477 1.622 3.782 2.68E-05 12 RS11171580 55975171 OR6C4 | OR10P1 C 0.2335 1.772 1.306 2.404 0.0002349 5 RS331706 124890054 ZNF608 | GRAMD3 A 0.1478 2.128 1.495 3.027 2.71E-05 1 RS821717 232059805 DISC1 C 0.2107 0.4419 0.2859 0.683 0.0002365 13 4 RS4834703 119691624 SEC24D A 0.09379 2.271 1.544 3.342 3.13E-05 13 RS17424088 23087983 FGF9 | LOC646201 A 0.1055 2.068 1.404 3.046 0.0002365 4 RS4302525 6552201 PPP2R2C | MAN2B2 A 0.1457 2.039 1.458 2.853 3.18E-05 2 RS17027173 103057043 IL18RAP A 0.2156 1.814 1.32 2.493 0.0002383 14 13 RS9547679 37402440 RFXAP A 0.07962 2.63 1.662 4.162 3.63E-05 16 RS16956168 56249774 DKFZP434H168 | GNAO1 A 0.1738 1.887 1.345 2.649 0.0002397 1 RS12723891 217143384 ESRRG G 0.1103 2.126 1.479 3.056 4.62E-05 13 RS7332913 79933284 RBM26 G 0.4909 0.5783 0.4317 0.7748 0.0002424 15 19 RS3745752 48339300 CRXFor Gpeer 0.4011 1.844 1.373 2.476 review4.84E-05 13 RS2763810 79979679 RBM26onlyA 0.4906 0.5783 0.4317 0.7748 0.0002424 22 RS138195 46151880 ATXN10 G 0.371 1.819 1.362 2.429 5.04E-05 7 RS2700976 37400570 ELMO1 A 0.2302 1.756 1.3 2.373 0.0002431 16 6 RS2876669 148216298 SAMD5 | SASH1 A 0.2328 1.891 1.389 2.573 5.12E-05 14 RS8009218 100057672 CCDC85C G 0.4541 1.733 1.292 2.325 0.0002438 4 RS4631029 21796070 KCNIP4 | NCRNA00099 G 0.09885 2.214 1.507 3.253 5.14E-05 2 RS17015041 78416572 SNAR-H | REG3G G 0.1397 1.975 1.373 2.841 0.0002458 17 20 RS6067635 49698104 MULTIPLE_GENES:3755,100128598 A 0.2601 1.903 1.393 2.6 5.24E-05 22 RS6007696 47968450 LOC100289420 | FAM19A5 G 0.1765 1.878 1.341 2.631 0.0002459 16 RS12920403 19756059 IQCK G 0.1235 2.109 1.464 3.036 6.06E-05 9 RS10809097 1057797 DMRT2 | SMARCA2 G 0.1373 1.999 1.38 2.895 0.0002483 18 6 RS9405142 1632299 GMDS G 0.01484 5.575 2.4 12.95 6.46E-05 15 RS11858577 67066563 SMAD6 A 0.07422 2.3 1.473 3.591 0.0002506 9 RS12115781 112672877 MULTIPLE_GENES:445815,114299 C 0.08401 2.359 1.545 3.601 6.97E-05 13 RS2147568 79991226 RBM26 | NDFIP2 G 0.4906 0.5797 0.4329 0.7762 0.0002516 19 9 RS7042036 119766969 ASTN2 G 0.3168 1.81 1.351 2.425 7.13E-05 10 RS7910400 6310176 PFKFB3 | LOC399715 G 0.4224 0.5523 0.4018 0.7591 0.000254 19 RS11668505 48348363 CRX | SULT2A1 G 0.3984 1.803 1.347 2.414 7.43E-05 2 RS3828312 100184719 AFF3 A 0.2045 1.82 1.32 2.508 0.0002564 7 RS10270298 104210910 LHFPL3 A 0.04825 3.011 1.743 5.2 7.76E-05 12 RS10859508 93919441 MRPL42 | SOCS2 A 0.3552 0.5428 0.3911 0.7533 0.0002576 20 14 RS10140464 49210639 MDGA2 | RPS29 G 0.198 1.894 1.38 2.6 7.82E-05 4 RS6822035 119710421 SEC24D A 0.2955 1.734 1.29 2.329 0.0002593 6 RS9352851 81392952 BCKDHB | FAM46A A 0.3728 1.787 1.34 2.384 7.82E-05 9 RS7855908 71517082 PIP5K1B G 0.05466 2.712 1.588 4.631 0.0002598 21 7 RS10953524 106618074 PIK3CG | PRKAR2B G 0.499 0.5472 0.4056 0.7381 7.86E-05 7 RS2723520 17813677 AHR | SNX13 G 0.4521 1.719 1.285 2.299 0.0002609 21 RS928873 43293371 PRDM15 A 0.278 0.4756 0.3287 0.6881 8.05E-05 3 RS967331 61447737 FHIT | PTPRG A 0.1559 1.902 1.347 2.686 0.0002631 22 7 RS2299146 42033195 GLI3 A 0.302 1.763 1.33 2.337 8.10E-05 7 RS11763025 21920165 DNAH11 A 0.3502 1.692 1.275 2.244 0.0002649 8 RS6586870 19709775 INTS10 | LPL A 0.3934 1.8 1.344 2.412 8.11E-05 20 RS2425847 44907927 CDH22 | SLC35C2 G 0.3856 1.706 1.28 2.274 0.0002664 23 1 RS12122147 74838630 TNNI3K G 0.1004 2.181 1.48 3.214 8.14E-05 1 RS10489478 176566350 PAPPA2 A 0.1734 0.4052 0.2493 0.6586 0.0002668 9 RS457417 4855514 RCL1 C 0.1306 2.027 1.426 2.883 8.27E-05 10 RS1915613 66483216 LOC100287799 | ANXA2P3 A 0.248 0.4801 0.3235 0.7124 0.0002689 24 15 RS11638941 39481950 LOC100289563 | C15orf54 G 0.1424 2.008 1.418 2.843 8.47E-05 23 RS7051093 90319063 LOC100288789 | PABPC5 A 0.1842 0.2732 0.1359 0.5491 0.0002701 7 RS2058894 97296791 ACN9 | TAC1 A 0.0368 3.17 1.783 5.636 8.52E-05 6 RS6456672 25347733 LRRC16A C 0.01417 6.505 2.374 17.82 0.0002709 25 17 RS11077582 69911104 LOC124685 | SOX9 A 0.06309 2.456 1.568 3.845 8.62E-05 23 RS5940927 90309551 LOC100288789 | PABPC5 A 0.1831 0.2738 0.1362 0.5504 0.000277 8 RS13254568 11514144 BLK | GATA4 A 0.1732 1.978 1.406 2.783 8.91E-05 23 RS10522055 90307072 LOC100288789 | PABPC5 G 0.1831 0.2738 0.1362 0.5504 0.000277 26 20 RS2425810 44890910 CDH22 | SLC35C2 A 0.3924 1.787 1.335 2.391 9.38E-05 20 RS2425807 44881545 CDH22 | SLC35C2 G 0.4136 1.714 1.281 2.291 0.0002808 10 RS3847355 71725020 COL13A1 | H2AFY2 A 0.442 1.789 1.336 2.396 9.57E-05 8 RS4510897 72277458 EYA1 | MSC G 0.1552 1.898 1.343 2.682 0.0002813 27 5 RS10078649 179877533 GFPT2 | CNOT6 A 0.09818 2.174 1.471 3.212 9.80E-05 8 RS10808467 115320834 CSMD3 | TRPS1 A 0.4899 0.5774 0.4291 0.7768 0.0002854 21 RS4276100 43275896 LOC100288232 | PRDM15 G 0.2652 1.823 1.348 2.467 9.91E-05 22 RS5754826 34476759 LARGE | ISX A 0.2893 0.5043 0.3483 0.73 0.0002861 28 2 RS11686987 107128885 RGPD3 | LOC644604 A 0.03038 3.33 1.817 6.105 9.97E-05 9 RS752357 36744752 MELK | PAX5 G 0.3478 0.5406 0.3877 0.7538 0.0002875 16 RS8055711 7040933 A2BP1 A 0.2881 1.825 1.348 2.47 1.00E-04 1 RS10489477 176562608 PAPPA2 A 0.169 0.3777 0.2231 0.6396 0.0002909 29 2 RS4853374 78442014 SNAR-H | REG3G A 0.1353 2.076 1.436 3.002 0.0001033 9 RS11792612 17442000 CNTLN A 0.1798 1.812 1.314 2.5 0.0002932 19 RS7260368 57135195 ZNF71 G 0.4855 1.756 1.321 2.335 0.0001076 10 RS12780002 6308484 PFKFB3 | LOC399715 A 0.3342 0.5359 0.3821 0.7514 0.0002986 30 6 RS9361640 81392005 BCKDHB | FAM46A C 0.3013 1.793 1.333 2.412 0.0001118 8 RS6997580 115321807 CSMD3 | TRPS1 C 0.4882 0.5787 0.4301 0.7787 0.000305 10 RS2895524 12595147 CAMK1D A 0.3465 0.5135 0.3662 0.7201 0.000112 3 RS1284745 102324326 ZPLD1 | LOC100287880 G 0.3313 1.714 1.279 2.297 0.0003073 31 14 RS17112396 42397496 LRFN5 | FSCB A 0.01316 6.11 2.436 15.32 0.0001143 1 RS10801786 90595332 ZNF326 | BARHL2 C 0.1781 1.816 1.313 2.511 0.0003103 8 RS7839240 39746386 ADAM2 | IDO1 G 0.3998 1.763 1.322 2.353 0.0001152 13 RS609418 37417427 RFXAP | SMAD9 C 0.2551 1.731 1.284 2.333 0.0003138 32 4 RS2880099 156432368 LOC100287564 | GUCY1A3 A 0.1191 2.059 1.426 2.974 0.0001162 6 RS4708617 168389858 HGC6.3 | KIF25 A 0.1134 0.2382 0.1091 0.5198 0.0003142 2 RS4951964 31300544 GALNT14 A 0.02767 3.535 1.86 6.719 0.0001165 4 RS3108140 107334044 LOC100288276 A 0.3107 1.7 1.273 2.269 0.0003187 33 14 RS1242558 83235421 SEL1L | FLRT2 A 0.2912 1.785 1.329 2.398 0.0001186 2 RS1370631 78410008 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233 15 RS12913680 90098150 LOC283761 | C15orf42 A 0.1093 2.144 1.452 3.164 0.0001236 2 RS2901929 78428973 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233 http://bmjopen.bmj.com/ 2 RS2894595 227237862 KIAA1486 | IRS1 C 0.2895 1.77 1.322 2.37 0.0001245 2 RS2164850 78423757 SNAR-H | REG3G A 0.141 1.949 1.355 2.803 0.0003233 34 4 RS1398835 21849857 NCRNA00099 A 0.09987 2.12 1.444 3.113 0.0001262 16 RS4786590 5125365 ALG1 G 0.0135 6.207 2.291 16.81 0.0003296 9 RS7035935 112706482 PALM2 C 0.06748 2.47 1.555 3.923 0.0001278 14 RS4902165 63115447 FLJ43390 | KCNH5 G 0.1387 1.921 1.345 2.744 0.0003297 35 2 RS11686479 31309930 GALNT14 A 0.028 3.505 1.845 6.66 0.0001284 6 RS7753022 86176625 NT5E A 0.1397 1.975 1.362 2.864 0.0003344 7 RS6461168 2170759 MAD1L1 G 0.2756 1.766 1.318 2.365 0.0001368 21 RS2839401 43302079 PRDM15 | C2CD2 A 0.2402 0.4871 0.3288 0.7217 0.0003358 36 13 RS6563501 37408545 RFXAP | SMAD9 G 0.2257 1.827 1.34 2.491 0.0001385 9 RS11144511 78218231 OSTF1 | PCSK5 A 0.4501 1.71 1.275 2.293 0.0003408 12 RS11111722 104104227 STAB2 A 0.05162 2.752 1.634 4.633 0.0001398 11 RS12799462 13977738 FAR1 | SPON1 C 0.2034 0.4419 0.2825 0.6911 0.0003456 37 2 RS11685519 107122121 RGPD3 | LOC644604 A 0.03138 3.156 1.746 5.702 0.0001408 6 RS12192975 20431741 E2F3 A 0.05634 2.357 1.474 3.769 0.0003457 7 RS4346898 21951033 CDCA7L A 0.391 1.715 1.299 2.264 0.0001422 23 RS180484 150318158 HMGB3 | LOC100128688 G 0.3481 1.961 1.356 2.837 0.0003485 38 21 RS870531 43276783 LOC100288232 | PRDM15 A 0.3292 0.5185 0.3697 0.7273 0.0001424 10 RS161016 110039615 LOC100128304 | LOC645318 G 0.3381 0.5504 0.3967 0.7635 0.0003497 5 RS716866 168645723 SLIT3 G 0.4356 1.741 1.308 2.317 0.0001427 16 RS7188523 55313079 LOC100132339 | IRX6 A 0.04757 2.643 1.551 4.504 0.0003502 39 5 RS6595621 124989579 ZNF608 | GRAMD3 A 0.2375 1.858 1.35 2.556 0.0001429 12 RS2303970 74471340 TRHDE | LOC552889 A 0.4673 0.5823 0.4328 0.7833 0.0003512 1 RS1183044 209255814 PLXNA2 | LOC642587 A 0.4922 1.744 1.309 2.323 0.000146 5 RS297853 168628678 SLIT3 A 0.3367 1.683 1.265 2.239 0.0003513 40 7 RS4140805 7727101 RPA3 C 0.4214 1.747 1.309 2.33 0.0001501 4 RS1908125 134392080 PCDH10 | PABPC4L G 0.2068 1.757 1.29 2.394 0.0003531 5 RS10447192 8896731 LOC100128382 | SEMA5A A 0.4069 0.5487 0.4022 0.7486 0.0001528 7 RS2237429 42031224 GLI3 A 0.1883 1.791 1.301 2.466 0.0003571 41 2 RS858940 50945699 NRXN1 A 0.3836 1.743 1.307 2.324 0.0001535 8 RS6989246 2015119 MYOM2 A 0.09514 2.071 1.389 3.088 0.0003573 7 RS10263415 7724805 LOC729852 | RPA3 A 0.4244 1.755 1.312 2.349 0.000154 8 RS7820007 142731821 FLJ43860 | NCRNA00051 A 0.3921 1.695 1.269 2.264 0.0003578 42 10 RS17452064 11242471 CUGBP2 C 0.1088 2.059 1.416 2.995 0.0001571 6 RS6932311 25346376 LRRC16A A 0.01383 6.219 2.279 16.97 0.0003597 on October 1, 2021 by guest. Protected copyright. 15 RS8041127 90135196 C15orf42 A 0.1043 2.118 1.435 3.128 0.0001596 22 RS138204 46161875 ATXN10 A 0.3647 1.686 1.266 2.247 0.0003607 43 6 RS6456676 25350468 LRRC16A C 0.01181 6.973 2.543 19.12 0.0001614 12 RS2365432 74439038 TRHDE | LOC552889 A 0.4136 0.5733 0.4223 0.7783 0.0003619 7 RS3778973 2151178 MAD1L1 A 0.208 1.809 1.329 2.462 0.0001653 15 RS4777184 70007323 LOC145837 | C15orf50 A 0.365 1.704 1.271 2.284 0.0003629 44 13 RS7998663 37435150 SMAD9 G 0.07456 2.47 1.542 3.954 0.000167 7 RS10233681 71941439 CALN1 | TYW1B A 0.2554 0.4969 0.3382 0.73 0.000366 8 RS10086641 39754780 ADAM2 | IDO1 A 0.4035 1.738 1.303 2.318 0.0001693 4 RS6828577 119704501 SEC24D A 0.2939 1.713 1.274 2.302 0.0003672 45 4 RS1878943 9766888 LOC100128096 | DRD5 A 0.1926 1.848 1.342 2.546 0.0001704 7 RS17410662 96231614 FLJ42280 | SHFM1 A 0.1427 1.877 1.327 2.655 0.0003686 6 RS220702 46867717 GPR116 G 0.09885 2.161 1.446 3.231 0.0001718 6 RS744143 20418718 E2F3 A 0.07596 2.156 1.412 3.29 0.0003701 46 11 RS554341 128685761 FLI1 | KCNJ1 G 0.1334 1.946 1.374 2.754 0.0001746 20 RS6033188 11585276 JAG1 | LOC728573 G 0.1353 0.3323 0.1812 0.6095 0.0003707 20 RS7508996 40396739 CHD6 | PTPRT G 0.08603 2.294 1.487 3.539 0.000175 5 RS4298224 8894289 LOC100128382 | SEMA5A A 0.3698 0.5636 0.411 0.7729 0.0003729 47 10 RS2184035 19778072 C10orf112 A 0.04453 2.7 1.606 4.538 0.0001773 2 RS10179116 100174645 AFF3 A 0.3522 1.714 1.273 2.306 0.0003763 7 RS7802845 111723488 DOCK4 G 0.07794 2.318 1.493 3.598 0.0001797 12 RS10431428 93898858 MRPL42 | SOCS2 G 0.2831 1.769 1.292 2.422 0.0003775 12 RS1316899 55966101 OR6C4 | OR10P1 A 0.1619 1.844 1.339 2.541 0.0001813 15 RS2289486 90161355 C15orf42 G 0.09177 2.068 1.385 3.086 0.0003776 48 1 RS10454444 176649181 PAPPA2 G 0.1754 0.3867 0.2352 0.6359 0.0001814 7 RS1404508 96228434 FLJ42280 | SHFM1 A 0.1518 1.855 1.32 2.608 0.0003779 7 RS17482047 7755484 RPA3 G 0.08835 2.177 1.448 3.272 0.0001832 4 RS6854328 119683492 SEC24D A 0.2939 1.711 1.272 2.3 0.0003788 49 1 RS10913237 176630416 PAPPA2 A 0.1748 0.3866 0.2349 0.6362 0.0001846 2 RS13030183 79909989 CTNNA2 A 0.4467 1.681 1.262 2.239 0.0003823 15 RS4774964 57854873 CGNL1 | GCOM1 G 0.1859 0.4038 0.2508 0.65 0.0001892 9 RS10969620 30323457 LOC100289571 | LOC100288436 A 0.0948 1.983 1.358 2.894 0.0003898 50 12 RS3871519 93915922 MRPL42 | SOCS2 A 0.3526 0.5349 0.3851 0.7429 0.0001894 6 RS6568370 105845531 PREP C 0.1087 2.048 1.378 3.045 0.0003918 2 RS1595422 78434830 SNAR-H | REG3G A 0.1326 2.03 1.4 2.943 0.0001897 2 RS12469727 227301483 KIAA1486 | IRS1 A 0.1653 1.819 1.307 2.532 0.0003923 51 4 RS298988 119648399 SEC24D A 0.226 1.793 1.32 2.437 0.0001899 2 RS13422448 49743324 FSHR | NRXN1 C 0.3188 0.5325 0.3758 0.7545 0.0003934 4 RS1491152 130547852 C4orf33 | LOC100132483 G 0.3347 1.74 1.3 2.329 0.0001949 23 RS5941802 90288180 LOC100288789 | PABPC5 C 0.179 0.2835 0.1412 0.5693 0.000394 52 20 RS6022863 52556423 SUMO1P1 | BCAS1 A 0.4028 0.5521 0.4038 0.7548 0.0001965 16 RS899236 56352905 GNAO1 A 0.2095 1.789 1.297 2.469 0.0003944 8 RS4875192 3187176 CSMD1 A 0.03914 2.825 1.634 4.883 0.0001996 17 RS767234 69886832 LOC124685 | SOX9 A 0.2186 1.766 1.289 2.42 0.0003964 53 15 RS17738087 26905021 GABRB3 G 0.0975 2.117 1.425 3.145 0.0002022 20 RS3091869 45425918 LOC100127904 | EYA2 A 0.4325 1.665 1.256 2.207 0.0003971 3 RS17061897 60020954 FHIT A 0.0587 2.476 1.534 3.994 0.0002039 16 RS17720036 27139728 C16orf82 | G 0.1309 1.929 1.341 2.775 0.0003997 54 1 RS6577559 9238095 GPR157 | LOC727721 A 0.1231 1.985 1.382 2.85 0.000205 10 RS6593445 45359483 LOC100289168 | TMEM72 G 0.1501 1.854 1.317 2.609 0.0004009 55 56 57 58 59 22 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 2 RS1356799 140567666 LOC647012 | LRP1B C 0.1977 0.439 0.2783 0.6926 0.0004021 2 RS893745 189088249 LOC729141 G 0.4933 0.5933 0.4406 0.799 0.0005872 3 RS6770667 71873579 PROK2 | RYBP A 0.2903 1.686 1.262 2.251 0.0004067 6 RS6911476 26612363 ABT1 | ZNF322A A 0.02328 4.264 1.864 9.752 0.0005914 7 2 RS6708183 78384746 SNAR-H | REG3G A 0.3252 1.706 1.269 2.295 0.0004101 4 RS6858670 137258380 LOC100289626 | LOC646316 G 0.4656 0.598 0.4459 0.8019 0.0005918 4 RS11730104 4682267 STX18 | LOC100289434 C 0.4136 1.665 1.255 2.209 0.0004125 4 RS7688325 137260052 LOC100289626 | LOC646316 A 0.4656 0.598 0.4459 0.8019 0.0005918 8 RS1497194 115351857 CSMD3 | TRPS1 G 0.4899 0.5859 0.4355 0.7883 0.0004131 8 RS290623 115395451 CSMD3 | TRPS1 A 0.473 0.5957 0.4433 0.8006 0.0005939 8 20 RS6032661 44733811 NCOA5 | CD40 G 0.2208 0.4692 0.3083 0.7142 0.0004147 16 RS9889027 81341363 BCMO1 | GAN G 0.2362 1.703 1.257 2.308 0.0005983 20 RS6064024 52546889 SUMO1P1 | BCAS1 G 0.4265 1.682 1.26 2.244 0.0004162 16 RS8055894 21190092 TMEM159 A 0.1414 1.894 1.315 2.728 0.0006004 9 11 RS7926320 13782568 FAR1 | SPON1 A 0.388 1.664 1.254 2.207 0.0004164 7 RS1859107 21859436 DNAH11 C 0.1333 0.367 0.207 0.6506 0.0006012 15 RS8043317 27751786 GABRG3 A 0.05466 2.493 1.501 4.141 0.000418 9 RS2417753 109405667 LOC644620 G 0.1549 1.845 1.3 2.618 0.0006025 10 17 RS1126642 42989063 GFAP A 0.03552 2.741 1.565 4.8 0.0004211 19 RS296366 48374538 SULT2A1 A 0.238 1.728 1.264 2.361 0.0006055 10 RS317499 109995284 LOC100128304 | LOC645318 A 0.3357 0.5558 0.4009 0.7706 0.0004257 4 RS10516407 21883570 LOC100131836 | KCNIP4 A 0.1891 1.791 1.284 2.499 0.0006057 11 15 RS4777990 93018587 C15orf32 G 0.2817 1.715 1.27 2.316 0.0004304 10 RS3818540 97165811 SORBS1 A 0.1043 2.08 1.369 3.162 0.0006058 2 RS10495786 33559429 LTBP1 G 0.1073 1.974 1.352 2.884 0.0004312 10 RS10733970 58773025 ZWINT | IPMK A 0.02161 5.501 2.075 14.58 0.0006088 12 2 RS10200850 240234901 HDAC4 A 0.0652 2.22 1.424 3.462 0.0004344 1 RS6673391 30731688 LOC100288450 | MATN1 G 0.108 2.04 1.357 3.068 0.0006099 3 RS9868354 108840227 MORC1 | LOC100288721 G 0.02901 3.265 1.689 6.313 0.0004352 5 RS17192980 6977768 POLS | LOC442132 G 0.09959 1.951 1.331 2.859 0.0006113 13 2 RS1922351 35300334 MYADML | LOC100288911 G 0.3924 1.681 1.259 2.246 0.0004369 10 RS4750200 6308646 PFKFB3 | LOC399715 A 0.3404 0.556 0.3974 0.7779 0.0006118 16 RS2356417 71356135 FTSJD1 | CALB2 G 0.1265 1.927 1.337 2.777 0.0004381 8 RS987341 18009367 ASAH1 | NAT1 A 0.05908 2.282 1.423 3.658 0.000613 14 15 RS2028465 71659358 THSD4 C 0.2156 1.771 1.288 2.437 0.000441 16 RS7203768 73301302 LOC100288121 | LOC401859 G 0.3962 0.5831 0.4282 0.7939 0.0006145 2 RS888281 56507299 CCDC85A G 0.3252 1.662 1.252 2.208 0.0004457 18 RS9953631 39047265 LOC647946 | KC6 G 0.3239 0.5518 0.3926 0.7755 0.0006176 15 23 RS12559251 90342188 LOC100288789For | PABPC5 C 0.175peer0.2844 0.1409 0.5738 0.0004462review12 RS10770407 19087150 CAPZA3 | onlyPLEKHA5 G 0.1381 1.939 1.327 2.833 0.0006179 16 RS1424106 79216793 WWOX A 0.4686 0.5839 0.4324 0.7884 0.0004463 8 RS7819086 115272300 CSMD3 | TRPS1 A 0.427 1.645 1.237 2.188 0.0006201 16 2 RS17553946 155347161 GALNT13 | KCNJ3 A 0.2236 0.4806 0.3192 0.7236 0.000449 1 RS381115 153099345 SPRR2F | SPRR2C A 0.1171 1.965 1.335 2.892 0.0006205 10 RS12220134 97129606 SORBS1 A 0.06174 2.421 1.477 3.966 0.0004499 3 RS4927870 195820576 TFRC | FLJ25996 A 0.1191 1.975 1.338 2.917 0.0006207 17 2 RS16859880 220267737 DNPEP | LOC100288941 A 0.02126 4.044 1.853 8.828 0.0004507 12 RS4761520 93910308 MRPL42 | SOCS2 G 0.3067 0.5467 0.3869 0.7726 0.0006213 2 RS1922353 35302429 MYADML | LOC100288911 C 0.418 1.689 1.26 2.264 0.0004518 19 RS3786656 14506664 CD97 A 0.3364 1.645 1.237 2.188 0.0006225 18 21 RS2830247 27848998 CYYR1 G 0.07625 2.105 1.388 3.193 0.0004582 10 RS2171302 3815292 LOC100287172 A 0.1721 1.826 1.293 2.578 0.000624 9 RS7036825 16125443 C9orf93 | BNC2 C 0.02193 3.692 1.778 7.666 0.0004596 10 RS624964 11231831 CUGBP2 G 0.3161 1.687 1.25 2.276 0.0006241 19 20 RS2868714 44886189 CDH22 | SLC35C2 G 0.4879 1.676 1.255 2.237 0.0004617 2 RS7594456 212403434 ERBB4 A 0.3185 1.659 1.241 2.218 0.0006277 13 RS9516888 97931709 MBNL2 A 0.4308 1.645 1.245 2.174 0.0004621 8 RS3808606 39769375 ADAM2 | IDO1 A 0.4727 1.659 1.241 2.217 0.0006282 20 10 RS3932533 3925728 LOC100130652 | LOC727894 A 0.277 0.5138 0.3538 0.7461 0.000468 17 RS7215344 2430126 METT10D | PAFAH1B1 G 0.2505 0.5055 0.3418 0.7476 0.0006328 9 RS2503351 19182893 ADFP | LOC100288002 G 0.1694 1.826 1.303 2.558 0.0004689 17 RS7222013 66267691 SLC16A6 T 0.2498 1.737 1.265 2.384 0.0006334 21 19 RS10469470 17187518 MYO9B C 0.1276 0.3353 0.1817 0.6185 0.0004691 10 RS317490 110051491 LOC100128304 | LOC645318 G 0.3357 0.5664 0.4088 0.7848 0.0006345 13 RS12869943 38356247 TRPC4 G 0.2549 0.5119 0.3517 0.745 0.0004698 6 RS9463046 44802718 SUPT3H G 0.03981 2.735 1.536 4.872 0.0006349 22 2 RS328632 9247782 MBOAT2 | ASAP2 A 0.1694 0.3992 0.2386 0.6679 0.0004703 8 RS290592 115370512 CSMD3 | TRPS1 G 0.4835 0.5978 0.445 0.8031 0.0006351 7 RS2158499 88558769 C7orf62 | ZNF804B A 0.4163 0.5763 0.4231 0.7849 0.000472 11 RS1528663 13967222 FAR1 | SPON1 G 0.1916 0.4625 0.2971 0.7199 0.0006367 10 RS12766986 3335061 PITRM1 | LOC100287172 G 0.1842 1.764 1.283 2.424 0.0004726 11 RS1507526 13940539 FAR1 | SPON1 A 0.363 1.658 1.24 2.216 0.0006371 23 13 RS6491347 97930890 MBNL2 A 0.4372 1.641 1.243 2.167 0.0004732 8 RS4876325 115239083 CSMD3 | TRPS1 G 0.448 1.641 1.235 2.18 0.0006374 6 RS6910470 149552517 UST | MAP3K7IP2 A 0.05815 2.367 1.46 3.836 0.0004738 6 RS1333224 137801943 IFNGR1 | OLIG3 A 0.1134 1.968 1.334 2.902 0.0006387 24 7 RS2160138 7755797 RPA3 G 0.4565 1.676 1.255 2.239 0.0004741 15 RS8030303 61772892 RORA | VPS13C A 0.02328 3.421 1.689 6.93 0.0006395 12 RS1357241 74509618 TRHDE | LOC552889 C 0.4821 0.5933 0.4426 0.7952 0.0004769 15 RS4906904 27034085 GABRB3 | GABRA5 A 0.07119 2.198 1.399 3.455 0.0006402 25 22 RS929020 47971041 LOC100289420 | FAM19A5 A 0.17 1.836 1.306 2.583 0.0004774 1 RS563189 83284533 LPHN2 | TTLL7 A 0.4204 1.634 1.233 2.167 0.0006456 20 RS2425835 44901836 CDH22 | SLC35C2 G 0.4337 1.673 1.253 2.234 0.0004801 3 RS6437813 108734088 MORC1 A 0.03036 3.124 1.623 6.011 0.0006474 26 9 RS11143833 71433741 PIP5K1B A 0.01757 3.784 1.793 7.987 0.0004811 10 RS12570530 12547825 CAMK1D G 0.2901 0.5473 0.3871 0.7739 0.0006483 22 RS7284093 51121521 SHANK3 A 0.1492 1.876 1.318 2.671 0.000482 10 RS11591904 45346953 LOC100289168 | TMEM72 G 0.2783 1.678 1.246 2.26 0.0006578 27 18 RS2564500 57230260 CCBE1 G 0.3822 1.668 1.252 2.224 0.0004825 20 RS6127015 52540606 SUMO1P1 | BCAS1 A 0.4585 0.5941 0.4403 0.8017 0.0006598 3 RS212020 59988905 FHIT A 0.09615 2.096 1.383 3.176 0.0004827 19 RS182420 48372195 CRX | SULT2A1 G 0.2407 1.722 1.259 2.354 0.0006606 28 6 RS3869129 31410649 MICA | HCP5 A 0.2142 1.757 1.28 2.411 0.000485 10 RS6585876 86450858 LOC100131699 | GRID1 A 0.335 1.649 1.236 2.199 0.0006643 15 RS1520015 37429504 MEIS2 | TMCO5A A 0.2085 0.4596 0.2969 0.7114 0.0004863 12 RS1800159 57593894 LRP1 A 0.3188 1.646 1.235 2.193 0.0006646 29 6 RS17800315 150765510 IYD | PLEKHG1 A 0.2419 1.709 1.265 2.309 0.0004875 6 RS3101826 160584853 LOC100289162 | SLC22A2 A 0.3333 1.634 1.232 2.168 0.0006649 6 RS3899823 31410597 MICA | HCP5 A 0.2146 1.757 1.28 2.412 0.0004887 20 RS3092379 45432808 LOC100127904 | EYA2 G 0.4342 0.5947 0.4408 0.8023 0.0006705 30 6 RS11757075 150766316 IYD | PLEKHG1 A 0.2421 1.708 1.264 2.309 0.0004914 11 RS161138 110892082 LOC120364 | C11orf53 A 0.3991 0.5867 0.4314 0.7978 0.0006739 12 RS2607916 716509 NINJ2 | LOC100049716 G 0.2301 1.738 1.274 2.372 0.0004928 1 RS884940 2223866 SKI A 0.4298 1.628 1.229 2.155 0.0006749 31 15 RS8030720 90143567 C15orf42 A 0.1005 2.016 1.359 2.99 0.0004954 2 RS13392442 221100327 SLC4A3 | EPHA4 A 0.03374 2.84 1.556 5.183 0.0006756 12 RS4489787 48811100 ZNF641 | ANP32D G 0.1162 0.3285 0.1756 0.6147 0.0004976 7 RS10254402 104192398 LHFPL3 G 0.06849 2.295 1.422 3.706 0.0006758 32 14 RS2369319 96355334 TCL1A | C14orf132 A 0.1775 1.788 1.289 2.48 0.0004985 8 RS3802328 19297051 CSGALNACT1 G 0.3313 0.5617 0.4028 0.7833 0.0006762 20 RS1883838 44761896 CD40 | CDH22 A 0.4126 0.5742 0.4201 0.7848 0.0005009 5 RS256314 34617222 LOC729915 | LOC401180 A 0.03306 3.221 1.64 6.323 0.0006787 33 14 RS12897979 100056208 CCDC85C A 0.368 1.676 1.253 2.243 0.0005036 22 RS5754818 34464710 LARGE | ISX A 0.4217 0.5851 0.4295 0.7971 0.000679
15 RS6576617 27035375 GABRB3 | GABRA5 G 0.1053 2.004 1.354 2.965 0.0005056 2 RS3771166 102986222 IL18R1 A 0.3782 1.654 1.237 2.211 0.0006793 http://bmjopen.bmj.com/ 34 18 RS4797173 466351 COLEC12 A 0.2517 1.707 1.263 2.308 0.0005061 9 RS10816456 98451928 PTCH1 | LOC100287056 A 0.225 0.5024 0.3378 0.7473 0.0006798 12 RS1032725 32641200 BICD1 | FGD4 G 0.3109 1.676 1.253 2.242 0.0005065 17 RS1858990 59735821 NACA2 | BRIP1 A 0.1707 1.783 1.277 2.488 0.0006802 35 3 RS2194938 139731000 CLSTN2 A 0.1955 1.764 1.281 2.43 0.0005077 7 RS10272276 41411165 C7orf10 | INHBA G 0.198 1.75 1.267 2.417 0.000681 12 RS1694747 17617111 LOC390298 | RERGL A 0.08632 2.06 1.37 3.095 0.0005083 14 RS8017842 100055722 CCDC85C G 0.4049 1.657 1.238 2.219 0.0006851 36 3 RS12497400 59986922 FHIT G 0.1193 0.3111 0.161 0.6012 0.0005136 17 RS9910870 1471611 PITPNA | LOC100288617 A 0.2534 0.5119 0.3478 0.7535 0.0006854 18 RS4797604 1213560 LOC100130247 | C18orf2 G 0.1215 2.024 1.359 3.012 0.0005148 9 RS7873155 16118758 C9orf93 | BNC2 C 0.02868 3.443 1.686 7.033 0.0006913 37 3 RS575653 187596668 BCL6 | LOC339929 A 0.2831 1.69 1.257 2.272 0.0005167 6 RS16895923 42633368 UBR2 G 0.1616 1.794 1.28 2.515 0.000692 17 RS7502216 36612948 SOCS7 | ARHGAP23 A 0.3826 0.5652 0.4095 0.7801 0.0005191 4 RS716051 113964288 ANK2 G 0.1614 1.851 1.297 2.642 0.0006941 2 RS11888120 29322384 C2orf71 | CLIP4 A 0.2456 1.707 1.262 2.309 0.0005198 18 RS1019868 66753605 CCDC102B | DOK6 C 0.01284 19.66 3.516 109.9 0.0006952 38 2 RS2390165 166059919 SCN3A A 0.448 1.673 1.251 2.238 0.0005199 2 RS10204137 102968212 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 5 RS10041897 169893751 KCNMB1 | KCNIP1 A 0.225 0.4793 0.3164 0.7261 0.0005201 2 RS10192157 102968356 IL1RL1 A 0.3802 1.652 1.236 2.209 0.0006956 39 20 RS3092167 44922717 CDH22 | SLC35C2 A 0.4578 1.646 1.242 2.182 0.0005235 2 RS10206753 102968362 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 18 RS3748415 10471732 LOC100288932 / APCDD1 A 0.1144 0.2917 0.1453 0.5853 0.0005263 9 RS10901213 133247453 HMCN2 C 0.113 1.931 1.32 2.825 0.0006968 40 5 RS6891322 169897753 KCNMB1 | KCNIP1 A 0.1781 0.4204 0.2575 0.6862 0.0005278 16 RS9924119 71343397 FTSJD1 | CALB2 G 0.1184 1.915 1.315 2.788 0.0006983 7 RS17170988 37361023 ELMO1 C 0.2196 0.4803 0.3173 0.7272 0.0005287 15 RS4632084 70049424 LOC145837 | C15orf50 A 0.281 0.535 0.3726 0.7681 0.0006989 41 6 RS2223451 150767547 IYD | PLEKHG1 A 0.2422 1.701 1.259 2.296 0.0005304 6 RS9497912 148296513 SAMD5 | SASH1 A 0.1269 1.901 1.311 2.756 0.0007016
20 RS2425856 44911954 CDH22 | SLC35C2 A 0.4416 1.641 1.24 2.171 0.0005317 2 RS831010 170139346 LRP2 G 0.165 1.792 1.279 2.511 0.0007023 on October 1, 2021 by guest. Protected copyright. 42 4 RS1438217 118039852 TRAM1L1 | LOC100288955 A 0.4335 1.652 1.243 2.194 0.0005318 12 RS7314861 29238335 CCDC91 | FAR2 A 0.1987 1.777 1.274 2.479 0.0007045 12 RS10771958 32638162 BICD1 | FGD4 C 0.3107 1.667 1.248 2.226 0.0005321 16 RS4287581 7032053 A2BP1 G 0.2594 1.703 1.252 2.318 0.0007073 43 7 RS2058893 97283118 ACN9 | TAC1 A 0.113 1.882 1.316 2.691 0.0005331 10 RS17571397 12572613 CAMK1D A 0.2601 0.5233 0.3595 0.7616 0.0007182 3 RS7626013 59961159 FHIT G 0.1076 2.046 1.364 3.069 0.0005428 10 RS4021187 126756083 CTBP2 A 0.4825 0.6061 0.4535 0.8102 0.0007191 44 9 RS1972384 110210248 RAD23B | KLF4 A 0.1113 1.977 1.343 2.91 0.0005493 2 RS2048765 240208154 HDAC4 G 0.0641 2.183 1.388 3.433 0.0007214 15 RS28408948 81998391 TMC3 | MEX3B G 0.0105 7.914 2.448 25.58 0.0005494 2 RS4676312 107658016 LOC649489 | LOC729121 C 0.05162 2.39 1.442 3.961 0.0007214 45 6 RS476235 160622637 LOC100289162 | SLC22A2 A 0.3428 1.65 1.242 2.193 0.0005521 8 RS925760 115227867 CSMD3 | TRPS1 A 0.4443 1.629 1.227 2.161 0.0007231 6 RS1885487 7668923 SNRNP48 | BMP6 G 0.3222 1.71 1.261 2.318 0.0005527 15 RS6494825 70048984 LOC145837 | C15orf50 A 0.4238 0.5973 0.443 0.8053 0.0007234 46 1 RS12061205 176554721 PAPPA2 A 0.1515 0.3932 0.2315 0.6681 0.0005568 17 RS1971043 20043515 CYTSB G 0.1838 0.4333 0.2668 0.7037 0.0007244 1 RS10494201 118252257 FAM46C | LOC100131261 A 0.3151 1.658 1.244 2.209 0.0005574 7 RS11971322 41386423 C7orf10 | INHBA A 0.1866 1.76 1.268 2.442 0.0007256 47 12 RS1992136 109777361 FOXN4 | MYO1H G 0.1103 0.3186 0.1664 0.6102 0.0005604 12 RS7305820 74550782 TRHDE | LOC552889 G 0.4953 1.635 1.229 2.175 0.0007274 7 RS10486908 82860532 PCLO | SEMA3E A 0.05803 2.398 1.459 3.942 0.0005634 10 RS11188295 97125874 SORBS1 A 0.08907 2.149 1.379 3.35 0.0007291 48 2 RS6744657 155012888 GALNT13 A 0.237 0.4909 0.3276 0.7357 0.0005662 3 RS3772460 59943532 FHIT G 0.1086 2.005 1.339 3.003 0.0007339 12 RS10861433 105877556 C12orf75 | NUAK1 A 0.226 0.4811 0.3173 0.7294 0.000568 9 RS295263 4840063 RCL1 G 0.1208 1.869 1.3 2.686 0.0007348 49 3 RS9860107 60040105 FHIT G 0.07996 2.157 1.393 3.34 0.0005682 18 RS11080417 10423834 VAPA | APCDD1 G 0.2679 1.698 1.249 2.309 0.000736 9 RS11103417 137494787 RXRA | COL5A1 G 0.2436 0.5026 0.3398 0.7432 0.0005683 17 RS7220740 59740793 NACA2 | BRIP1 A 0.3133 1.641 1.231 2.188 0.0007379 50 4 RS2196716 118055429 TRAM1L1 | LOC100288955 C 0.4764 0.5955 0.4434 0.7997 0.0005697 2 RS6431565 238836683 RAMP1 | UBE2F A 0.3316 1.67 1.24 2.249 0.0007438 15 RS1435835 26412976 LOC100128714 | GABRB3 G 0.1292 1.931 1.328 2.808 0.0005763 12 RS12823898 93860565 UBE2N | MRPL42 A 0.114 1.978 1.331 2.939 0.0007443 51 21 RS4919983 43301781 PRDM15 | C2CD2 A 0.3155 0.5512 0.3926 0.7737 0.0005765 4 RS1395114 113953581 ANK2 A 0.191 1.797 1.278 2.526 0.0007508 1 RS12067235 205281255 NUAK2 A 0.2213 0.4798 0.3158 0.7291 0.0005813 11 RS627354 74028944 P4HA3 | PGM2L1 A 0.4325 1.621 1.224 2.146 0.0007512 52 8 RS6985983 115289224 CSMD3 | TRPS1 A 0.4265 1.649 1.24 2.193 0.0005824 1 RS10462021 7897133 PER3 G 0.1842 1.744 1.262 2.411 0.0007523 8 RS1552237 87045760 REXO1L2P | PSKH2 A 0.2854 1.672 1.247 2.241 0.0005854 1 RS12125785 209254017 PLXNA2 | LOC642587 A 0.4015 0.595 0.4398 0.8049 0.0007573 9 RS10739231 98453680 PTCH1 | LOC100287056 A 0.4281 0.5842 0.43 0.7936 0.0005855 13 RS9531028 80767832 LOC729479 | SPRY2 A 0.1899 0.4483 0.2811 0.7151 0.0007574 53 12 RS4767252 115187724 TBX3 | MED13L A 0.33 1.67 1.247 2.237 0.0005858 19 RS11083726 29388987 LOC148189 | LOC148145 A 0.4565 1.645 1.231 2.198 0.0007601 12 RS995603 29235933 CCDC91 | FAR2 C 0.1993 1.785 1.283 2.484 0.0005862 7 RS4730427 78285960 MAGI2 A 0.1309 0.3623 0.2006 0.6542 0.0007604 54 7 RS206196 20361873 MACC1 | ITGB8 A 0.4213 0.5846 0.4305 0.7939 0.0005865 1 RS7527871 2281726 MORN1 | LOC100129534 C 0.1715 1.769 1.269 2.465 0.0007626 55 56 57 58 59 23 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 12 RS11046362 22460338 ST8SIA1 A 0.1046 2.003 1.336 3.001 0.000764 5 RS31036 146615260 STK32A C 0.3286 1.644 1.227 2.203 0.0008802 7 RS2204779 118438184 ANKRD7 | LOC100287273 G 0.01619 4.926 1.946 12.47 0.0007653 7 RS974060 97271508 ACN9 | TAC1 G 0.2939 1.643 1.226 2.201 0.0008811 8 RS1606891 115206234 CSMD3 | TRPS1 A 0.4483 1.627 1.225 2.16 0.0007696 2 RS10183733 135071570 MGAT5 A 0.1772 1.773 1.265 2.484 0.0008825 7 12 RS12368128 125648067 AACS | TMEM132B A 0.2449 0.5034 0.3375 0.751 0.0007698 18 RS4306606 55334295 ATP8B1 A 0.2713 0.5341 0.3691 0.773 0.0008836 4 RS7656244 65170415 SRD5A2L2 A 0.1859 0.4634 0.2959 0.7255 0.0007708 18 RS2345 75279459 GALR1 | LOC100132713 G 0.06883 2.148 1.368 3.372 0.0008951 8 2 RS6547192 78434082 SNAR-H | REG3G A 0.1373 1.89 1.304 2.739 0.0007725 3 RS4974500 134109630 AMOTL2 | ANAPC13 G 0.4943 0.6023 0.4466 0.8124 0.0008961 2 RS6432964 149986664 LYPD6B G 0.3654 0.5723 0.4133 0.7924 0.0007749 3 RS1879784 183203034 MCF2L2 | KLHL6 A 0.4757 1.643 1.226 2.202 0.000898 9 8 RS167069 115423582 CSMD3 | TRPS1 G 0.4848 0.6037 0.4497 0.8102 0.0007753 4 RS12511494 113770535 ANK2 A 0.141 1.859 1.289 2.682 0.0009018 8 RS10283337 1319724 LOC401442 | DLGAP2 G 0.4052 1.625 1.224 2.157 0.0007815 6 RS2474880 63460047 KHDRBS2 | LOC100128610 A 0.4325 1.638 1.224 2.192 0.0009037 10 4 RS7676913 118043203 TRAM1L1 | LOC100288955 C 0.4706 0.6035 0.4495 0.8104 0.0007854 15 RS11636337 70044354 LOC145837 | C15orf50 G 0.251 0.5284 0.3625 0.7704 0.0009116 1 RS493631 153143048 SPRR2G | LELP1 A 0.1195 1.968 1.325 2.921 0.0007874 11 RS1852755 13996686 SPON1 G 0.2611 0.529 0.3631 0.7707 0.000912 11 14 RS878554 94606179 IFI27L2 | PPP4R4 A 0.2057 1.746 1.261 2.418 0.0007884 9 RS16933086 36215581 GNE A 0.2016 1.685 1.238 2.293 0.0009121 11 RS10894162 129781016 PRDM10 A 0.02092 3.344 1.652 6.767 0.000792 8 RS12547514 55806439 LOC100287651 | XKR4 C 0.4172 1.638 1.224 2.194 0.0009135 12 5 RS1154758 162716850 GABRG2 | CCNG1 G 0.4443 0.5973 0.4421 0.8071 0.0007932 7 RS13237260 7779091 RPA3 | LOC729852 A 0.2415 1.665 1.232 2.252 0.000918 10 RS7914288 71739556 COL13A1 | H2AFY2 G 0.389 1.629 1.225 2.167 0.0007965 13 RS9532283 39394277 FREM2 A 0.4673 1.626 1.22 2.168 0.0009189 13 17 RS7219555 59740122 NACA2 | BRIP1 C 0.3151 1.636 1.227 2.181 0.0007979 2 RS12619691 35306866 MYADML | LOC100288911 A 0.3937 1.628 1.22 2.172 0.0009197 12 RS10783915 59001589 XRCC6BP1 | LRIG3 G 0.332 0.5723 0.413 0.793 0.000798 3 RS6766938 67150565 KBTBD8 | SUCLG2 A 0.2142 1.721 1.248 2.373 0.0009197 14 5 RS951087 13328289 CTNND2 | DNAH5 A 0.1107 1.971 1.325 2.931 0.0008055 11 RS1852757 13987892 SPON1 G 0.1866 0.4711 0.3018 0.7353 0.0009207 13 RS9510573 23680990 LOC646201 | SGCG A 0.1889 1.746 1.26 2.419 0.0008068 9 RS7846713 8079286 C9orf123 | PTPRD A 0.2375 1.7 1.242 2.327 0.0009219 15 14 RS885845 96729728 BDKRB1 ForA 0.37peer1.639 1.228 2.189 0.0008088review5 RS457638 80359968 RASGRF2 onlyG 0.1867 0.468 0.2986 0.7333 0.0009227 1 RS6676869 209249614 PLXNA2 | LOC642587 G 0.3036 0.5568 0.3952 0.7844 0.0008115 19 RS9917042 14494473 CD97 G 0.3212 1.623 1.219 2.162 0.0009241 16 6 RS9283918 53926867 C6orf142 A 0.05904 2.267 1.404 3.661 0.0008121 8 RS12677088 19309945 CSGALNACT1 A 0.3126 0.5637 0.4015 0.7913 0.0009254 6 RS9296737 53926716 C6orf142 G 0.05904 2.267 1.404 3.661 0.0008121 6 RS6570859 148829305 SASH1 G 0.3333 0.5748 0.4141 0.7977 0.0009283 17 3 RS9841113 116455714 LOC285194 | IGSF11 A 0.4339 1.62 1.222 2.149 0.0008144 15 RS7183427 39503508 LOC100289563 | C15orf54 C 0.1835 1.766 1.261 2.472 0.0009291 5 RS1023617 172740329 NKX2-5 | STC2 A 0.3128 1.66 1.234 2.235 0.0008162 10 RS11003908 55645842 PCDH15 A 0.02092 4.64 1.87 11.51 0.0009301 18 3 RS4076927 46730749 ALS2CL G 0.3171 0.55 0.3875 0.7807 0.0008215 14 RS1197382 81146776 C14orf145 G 0.08974 0.2693 0.1238 0.5856 0.0009328 8 RS10091115 62684067 LOC100287835 | NKAIN3 C 0.3381 0.5715 0.4117 0.7934 0.0008288 7 RS3800895 2125942 MAD1L1 A 0.2743 1.636 1.222 2.19 0.0009334 19 12 RS2364227 93893510 MRPL42 G 0.303 0.5558 0.3939 0.7844 0.0008315 10 RS7905027 124114432 BTBD16 | PLEKHA1 G 0.3932 0.5897 0.4313 0.8063 0.0009347 21 RS2832279 30613466 C21orf109 A 0.3323 0.5744 0.415 0.7951 0.0008317 2 RS6748959 175424127 GPR155 | WIPF1 G 0.4906 0.6052 0.4495 0.8148 0.0009351 20 4 RS12508492 112696927 LOC729065 | C4orf32 G 0.05331 2.324 1.417 3.81 0.0008328 9 RS10814369 36292975 GNE | LOC646993 A 0.192 1.714 1.246 2.359 0.0009351 6 RS6940801 169655667 THBS2 | LOC100289536 G 0.1498 1.779 1.269 2.493 0.0008332 11 RS12793784 26700112 SLC5A12 A 0.1208 1.926 1.306 2.84 0.0009361 21 3 RS17485052 138638081 PIK3CB | FOXL2 A 0.1991 1.75 1.26 2.43 0.0008398 16 RS6564372 73324074 LOC100288121 | LOC401859 A 0.4986 0.6197 0.4668 0.8228 0.0009375 12 RS10746042 106111254 C12orf75 | NUAK1 G 0.2399 1.695 1.244 2.311 0.0008398 2 RS7559847 222068646 SLC4A3 | EPHA4 A 0.385 1.639 1.223 2.197 0.0009419 22 22 RS138154 46107105 ATXN10 G 0.2319 1.673 1.237 2.263 0.0008421 3 RS9790156 139814577 CLSTN2 A 0.1002 1.978 1.32 2.964 0.0009425 1 RS12120640 74821837 TNNI3K G 0.1205 1.86 1.292 2.678 0.0008424 1 RS1925421 39250964 LOC400750 | RRAGC A 0.2567 1.645 1.225 2.21 0.0009447 23 20 RS2425817 44895286 CDH22 | SLC35C2 A 0.4329 1.633 1.225 2.179 0.0008438 23 RS10521479 55583383 LOC644893 | FOXR2 A 0.1163 2.228 1.386 3.582 0.0009464 7 RS4730197 106446900 FLJ36031 | PIK3CG C 0.1323 0.3697 0.2061 0.6631 0.0008448 5 RS7446493 6974432 POLS | LOC442132 G 0.1029 1.905 1.3 2.791 0.0009471 6 RS2817457 156943434 NOX3 | ARID1B A 0.4774 1.648 1.229 2.21 0.0008462 1 RS3907135 106526262 LOC642337 | LOC126987 G 0.4926 1.628 1.219 2.173 0.0009488 24 1 RS3006993 114570444 OLFML3 | SYT6 A 0.2696 1.671 1.236 2.258 0.0008466 2 RS4588165 35326722 MYADML | LOC100288911 A 0.414 1.631 1.22 2.18 0.0009507 11 RS7128466 120491438 ARHGEF12 | GRIK4 A 0.1437 0.4015 0.2349 0.6863 0.000849 5 RS2401793 13332878 CTNND2 | DNAH5 C 0.1123 1.952 1.313 2.904 0.0009537 25 11 RS11025673 20663967 SLC6A5 G 0.3812 1.6 1.214 2.109 0.000851 3 RS11709505 53749498 CACNA1D G 0.2328 1.684 1.236 2.294 0.0009593 3 RS1523759 77604242 ROBO2 G 0.1711 0.4267 0.2586 0.7039 0.0008541 2 RS9288433 212407077 ERBB4 A 0.3101 1.634 1.221 2.187 0.000961 26 12 RS10746041 106108279 C12orf75 | NUAK1 G 0.2264 1.711 1.248 2.347 0.0008554 10 RS4751800 122636730 BRWD2 G 0.1559 0.4236 0.2543 0.7054 0.0009626 6 RS4895999 134147153 MGC34034 G 0.3805 1.614 1.218 2.139 0.0008575 23 RS6639393 4052872 LOC729162 | NLGN4X G 0.2684 1.885 1.294 2.747 0.0009636 27 3 RS10804937 103308760 LOC100287880 | ALCAM A 0.3184 1.624 1.221 2.16 0.0008578 20 RS6109068 11588817 JAG1 | LOC728573 G 0.1211 0.3592 0.1955 0.6599 0.0009683 13 RS9557607 101950392 NALCN G 0.03036 2.92 1.555 5.484 0.0008589 7 RS7788775 141817070 LOC93432 A 0.2318 0.5088 0.3406 0.7601 0.0009702 28 10 RS10826503 29010588 BAMBI | LOC729601 A 0.3866 1.636 1.225 2.185 0.0008601 15 RS1426079 70044495 LOC145837 | C15orf50 A 0.4123 1.63 1.219 2.178 0.0009716 10 RS619030 11206281 CUGBP2 A 0.2328 1.712 1.248 2.349 0.0008633 7 RS2613585 111186693 IMMP2L G 0.1164 0.317 0.1601 0.6274 0.0009726 29 18 RS151472 3447871 TGIF1 A 0.2908 1.652 1.229 2.219 0.0008653 9 RS1889321 113300835 SVEP1 G 0.2682 1.65 1.225 2.221 0.0009738 1 RS476093 181787259 CACNA1E | ZNF648 C 0.2446 1.683 1.239 2.285 0.0008663 10 RS7094645 72796318 PCBD1 | UNC5B A 0.2481 0.5231 0.3559 0.7689 0.0009773 30 1 RS12130076 240918105 GREM2 | LOC645939 G 0.1238 0.3612 0.1983 0.6577 0.0008678 3 RS7633016 46728659 ALS2CL A 0.3192 0.5552 0.3913 0.7877 0.0009774 1 RS2169089 5088617 AJAP1 | LOC100287877 A 0.2328 0.5191 0.3529 0.7635 0.0008679 18 RS6506772 75918499 GALR1 | LOC100132713 G 0.254 0.5182 0.3505 0.7662 0.000983 31 20 RS6032678 44777295 CD40 | CDH22 A 0.4514 1.631 1.223 2.176 0.000871 3 RS13084580 39188182 CSRNP1 A 0.1111 1.906 1.299 2.798 0.000987 12 RS4766019 3089354 TEAD4 A 0.123 0.3318 0.1733 0.6353 0.0008719 7 RS10276017 140979971 LOC100129514 | LOC100130169 A 0.04082 2.507 1.451 4.333 0.0009892 32 2 RS16985278 18734072 KCNS3 | RDH14 A 0.2635 1.684 1.239 2.288 0.0008739 13 RS7323413 55363417 LOC100287887 | PRR20 C 0.03745 3.145 1.59 6.221 0.0009912 3 RS1163368 108946466 C3orf66 | DPPA2 A 0.3219 1.621 1.22 2.155 0.0008751 9 RS6415721 14964952 FREM1 | LOC389705 G 0.319 0.571 0.409 0.797 0.0009913 33 7 RS11971933 111717031 DOCK4 G 0.09615 2.053 1.344 3.135 0.0008756 20 RS11086320 49687647 MULTIPLE_GENES:3755,100128598 G 0.3013 1.662 1.228 2.249 0.0009953
18 RS7506330 9177894 ANKRD12 G 0.4774 0.6034 0.4481 0.8125 0.0008765 4 RS1995917 24207765 PPARGC1A | LOC729175 G 0.06545 2.127 1.357 3.333 0.0009991 http://bmjopen.bmj.com/ 34 6 RS9442693 72386001 C6orf155 | RIMS1 A 0.05027 2.582 1.476 4.514 0.0008798 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 24 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 C. SUPPLEMENTAL REFERENCES 3 4 1. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components 5 analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38:904- 6 909 7 2. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by 8 genome-wide association. Genomics. 2008;92:265-272 9 3. Liu KY, Muehlschlegel JD, Perry TE, Fox AA, Collard CD, Body SC, Shernan SK. Common 10 genetic variants on chromosome 9p21 predict perioperative myocardial injury after coronary 11 artery bypass graft surgery. J Thorac Cardiovasc Surg. 2010;139:483-488, 488 e481-482 12 4. Muehlschlegel JD, Liu KY, Perry TE, Fox AA, Collard CD, Shernan SK, Body SC. Chromosome 13 14 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation. 15 2010;122:S60For-65 peer review only 16 5. Kraft P. Efficient two-stage genome-wide association designs based on false positive report 17 probabilities. Pac Symp Biocomput. 2006:523-534 18 6. Zuo Y, Zou G, Wang J, Zhao H, Liang H. Optimal two-stage design for case-control association 19 analysis incorporating genotyping errors. Ann Hum Genet. 2008;72:375-387 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 25 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 STROBE Statement—checklist of items that should be included in reports of observational studies 3 4 Item 5 No Recommendation 6 Title and abstract 7 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 8 (b) Provide in the abstract an informative and balanced summary of what was done 9 and what was found 10 Introduction 11 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 13 Objectives 3 State specific objectives, including any prespecified hypotheses 14 Methods 15 For peer review only 16 Study design 4 Present key elements of study design early in the paper 17 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 18 exposure, follow-up, and data collection 19 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of 20 21 selection of participants. Describe methods of follow-up 22 Case-control study—Give the eligibility criteria, and the sources and methods of 23 case ascertainment and control selection. Give the rationale for the choice of cases 24 and controls 25 26 Cross-sectional study—Give the eligibility criteria, and the sources and methods of 27 selection of participants 28 (b) Cohort study—For matched studies, give matching criteria and number of 29 exposed and unexposed 30 Case-control study—For matched studies, give matching criteria and the number of 31 32 controls per case 33 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect http://bmjopen.bmj.com/ 34 modifiers. Give diagnostic criteria, if applicable 35 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 36 37 measurement assessment (measurement). Describe comparability of assessment methods if there 38 is more than one group 39 Bias 9 Describe any efforts to address potential sources of bias 40 Study size 10 Explain how the study size was arrived at 41
42 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, on October 1, 2021 by guest. Protected copyright. 43 describe which groupings were chosen and why 44 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 45 (b) Describe any methods used to examine subgroups and interactions 46 (c) Explain how missing data were addressed 47 48 (d) Cohort study—If applicable, explain how loss to follow-up was addressed 49 Case-control study—If applicable, explain how matching of cases and controls was 50 addressed 51 Cross-sectional study—If applicable, describe analytical methods taking account of 52 53 sampling strategy 54 (e) Describe any sensitivity analyses 55 Continued on next page 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 52 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Results 4 5 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, 6 examined for eligibility, confirmed eligible, included in the study, completing follow-up, and 7 analysed 8 (b) Give reasons for non-participation at each stage 9 (c) Consider use of a flow diagram 10 11 Descriptive 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information 12 data on exposures and potential confounders 13 (b) Indicate number of participants with missing data for each variable of interest 14 (c) Cohort study—Summarise follow-up time (eg, average and total amount) 15 For peer review only 16 Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 17 Case-control study—Report numbers in each exposure category, or summary measures of 18 exposure 19 Cross-sectional study—Report numbers of outcome events or summary measures 20 21 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their 22 precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and 23 why they were included 24 (b) Report category boundaries when continuous variables were categorized 25 26 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful 27 time period 28 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 29 analyses 30 31 Discussion 32 Key results 18 Summarise key results with reference to study objectives 33 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. 34 http://bmjopen.bmj.com/ Discuss both direction and magnitude of any potential bias 35 36 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity 37 of analyses, results from similar studies, and other relevant evidence 38 Generalisability 21 Discuss the generalisability (external validity) of the study results 39 Other information 40 41 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable,
42 for the original study on which the present article is based on October 1, 2021 by guest. Protected copyright. 43 44 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and 45 46 unexposed groups in cohort and cross-sectional studies. 47 48 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 49 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 50 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 51 52 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 53 available at www.strobe-statement.org. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
Genome-wide association study of perioperative myocardial infarction after coronary artery bypass surgery
ForJournal: peerBMJ Open review only Manuscript ID: bmjopen-2014-006920.R2
Article Type: Research
Date Submitted by the Author: 10-Mar-2015
Complete List of Authors: Kertai, Miklos; Duke University, Department of Anesthesiology/Cardiothoracic Li, Yi-Ju; Duke University, Department of Biostatistics and Bioinformatics Li, Yen-Wei; Duke University, Department of Biostatistics and Bioinformatics Ji, Yunqi; Duke University, Department of Biostatistics and Bioinformatics Alexander, John; Duke University, Department of Medicine/Cardiology; Duke University, Duke Clinical Research Institute Newman, Mark; Duke University, Department of Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute Smith, Peter; Duke University, Department of Surgery/Cardiothoracic Joseph, Diane; Duke University, Duke Clinical Research Institute Mathew, Joseph; Duke University, Department of Anesthesiology/Cardiothoracic http://bmjopen.bmj.com/ Podgoreanu, Mihai; Duke University, Department of Anesthesiology/Cardiothoracic; Duke University, Duke Clinical Research Institute
Primary Subject Genetics and genomics Heading:
Secondary Subject Heading: Cardiovascular medicine, Surgery, Epidemiology on October 1, 2021 by guest. Protected copyright. Myocardial infarction < CARDIOLOGY, SURGERY, Adult intensive & critical Keywords: care < ANAESTHETICS, Anaesthesia in cardiology < ANAESTHETICS, Cardiac Epidemiology < CARDIOLOGY, GENETICS
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from ; Mark F. Newman, F. Newman, ;Mark 2,6 ; ; 1,6 ; John Alexander, MD, MPH MD, JohnAlexander, ; 4 Department of Biostatistics and Bioinformatics; Bioinformatics; of Biostatisticsand Department 4 ; Mihai V. Podgoreanu, MD Podgoreanu, V. ;Mihai 1 : 3,100
Cardiac Surgery; Surgery; Cardiac 3 BMJ Open , Ph.D.; Yunqi Ji, PhD Yunqi Ph.D.; , http://bmjopen.bmj.com/ 4 ; Joseph P. Mathew, MD, MPH MD, JosephP. Mathew, ; 7 Cardiology, and and Cardiology, ; Yen Wei Li ;Yen Wei 2 4,5 Duke Clinical Research Institute; Duke University, Durham, NC, USA USA NC, Durham, University, Duke Institute; Research Clinical Duke on October 1, 2021 by guest. Protected copyright. 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml ; Yi Ju Li, PhD ;Yi Ju Li, ; Diane Joseph, MA ;Diane 1 3 For peer review only cardiopulmonary bypass, genome wide association study, inflammation, myocardial infarction, surgery infarction, inflammation, myocardial study, association bypass, genome wide cardiopulmonary ; Peter K. Smith, MD Smith, K. ;Peter 1,6 Divisions of Cardiothoracic Anesthesiology, Anesthesiology, Cardiothoracic of Divisions Physiology Institute; Molecular Duke Duke University Medical Center, Durham, NC 27710. 27710. NC Durham, Center, Medical University Duke Email: [email protected] 919 681 4720; Telephone: 919 681 4776; Fax: Words: Key for the Duke Perioperative Genetics and Safety Outcomes (PEGASUS) Investigative Team Investigative (PEGASUS) Outcomes Safety and Genetics Perioperative Duke the for MD Genome-wide association study of perioperative myocardial infarction after coronary artery bypass surgery bypassartery surgery coronary after infarction myocardial perioperative study of association Genome-wide PhD MD, Kertai, MiklosD. 1 5 of Anesthesiology, MD; Department 3094, Box MihaiPodgoreanu, to V. Correspondence references) and tables, figures abstract, title page, (excluding Count Word Page 1 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 2 of 52 in stage II. A secondary secondary A instage II. 5 2 for rs609418), or a dichotomized phenotype based on CK on extreme based phenotype dichotomized a or rs609418), for 3 BMJ Open http://bmjopen.bmj.com/ The primary outcome variable was perioperative MI, defined as CK MB values CK MB as MI, defined perioperative variable was outcome The primary on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We conducted a stage I genome wide association study (GWAS) in 1,433 ethnically diverse patients of both genders both of genders patients diverse ethnically (GWAS)in1,433 study association stage a genome wide I conducted We Identification of patient populations at risk for developing myocardial infarction (MI), or alternatively, with intrinsic with or (MI), alternatively, infarction myocardial developing at for populations risk patient of Identification Following quality control and adjustment for clinical covariates, we identified 521 single nucleotide polymorphisms (SNPs) polymorphisms nucleotide single 521 weidentified covariates, clinical for and adjustment control quality Following ABSTRACT ABSTRACT Objectives:
analysis using CK MB as a quantitative trait (min p = 1.26x10 = p (min trait quantitative as CK MB a using analysis Participants: cardioprotection, or responsiveness to a protective agent remain high priority knowledge gaps. We sought to identify novel common novel toidentify sought We gaps. knowledge high priority remain agent protective toa responsiveness or cardioprotection, using genome wide bypass grafting artery coronary undergoing MI patients in perioperative with associated variants genetic methodology. association States. the United centers across surgery cardiac and tertiary secondary 107 Setting: ≥10xULN during the first postoperative day, and not attributable to preoperative MI. Secondary outcomes included postoperative CK postoperative included outcomes Secondary MI. preoperative to and attributable not day, first during the postoperative ≥10xULN distribution. CK MB the of quartiles extreme on based phenotype dichotomized a or trait, quantitative a as MB Results: (112 cases/1,321 controls) from the Genetics of Myocardial Adverse Outcomes and Graft Failure (GeneMAGIC) study, and a stage study, a stage II and (GeneMAGIC) Failure Graft and Adverse Outcomes Myocardial of Genetics the from controls) cases/1,321 (112 Duke and the GeneMAGIC from combined controls) cases/1,830 (225 patients 2,055 of inan expanded population analysis bypass coronary non emergent primary undergoing Patients studies. (PEGASUS) Outcomes and Safety Genetics Perioperative included. were grafting measures: and outcome secondary Primary in the stage I GWAS analysis. Among these, 8 common variants in 3 genes or intergenic regions met p < 10 < p regionsmet intergenic or genes in3 common these, 8 variants Among analysis. GWAS stage I inthe 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 3 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml for rs4834703) supported these findings. Pathway analysis revealed that genes harboring top scoring scoring top harboring that genes revealed analysis Pathway findings. these supported rs4834703) for 6 For peer review only Using 2 stage GWAS and pathway analysis, we identified and prioritized several potential susceptibility loci for loci potential susceptibility several and prioritized we identified analysis, pathway and GWAS 2 stage Using We conducted comprehensive and complementary single marker and pathway based genome wide association analyses. analyses. association pathway based and genome wide marker single complementary and comprehensive conducted We This is the first genome wide association study of perioperative myocardial infarction, using prospective cohorts of cardiac cardiac of cohorts prospective using myocardial of infarction, study perioperative association genome wide first is the This factors. risk non genetic and adjustment full phenotype, for primary the of definitions standard patients, surgical The study is powered to detect relatively large effect sizes. large effect relatively to is powered detect study The Predominantly Caucasian cohort, thus findings cannot be generalized to other populations. populations. toother cannot generalized be findings thus Caucasian cohort, Predominantly Rare genetic variant effects not analyzed. analyzed. not effects variant genetic Rare
• • • • • MB values (min p = 7.72x10 = p (min values MB
variants cluster in pathways of biologic relevance to extracellular matrix remodeling, ER to Golgi transport and inflammation. and transport ER to Golgi matrix remodeling, toextracellular of relevance biologic pathways in cluster variants Conclusions: perioperative MI. perioperative STUDY THIS OF LIMITATIONS AND STRENGTHS
Page 3 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 4 of 52
1 Thus, the overall influence of of commoninfluence overall the Thus, 8 12
2 4 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. However, although family based methods are not practical for studying perioperative MI, its genetic MI, its genetic perioperative for studying practical not methods are family based although However, 3 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Yet only a few studies have identified allelic associations with altered susceptibility to myocardial ischemia reperfusion ischemia reperfusion tomyocardial susceptibility altered with allelic associations identified few have a only studies Yet 7 The evidence for heritability of MI is striking, supported both by family studies and more recently, by a number of well of number a by recently, more studies and family by both supported ofis striking, MI heritability for evidence The Despite advances in surgical techniques and pharmacologic therapy, the incidence of myocardial infarction (MI) after coronary coronary after (MI) infarction of myocardial the incidence therapy, pharmacologic and techniques insurgical advances Despite individual genetic background may also significantly modulate the magnitude of postoperative inflammatory response after cardiac after cardiac response inflammatory postoperative magnitudeof the modulate significantly also backgroundmay genetic individual surgery. genetic variation on the incidence of perioperative MI remains poorly understood. understood. poorly remains MI of perioperative incidence the on variation genetic Identification of populations at risk for developing large infarcts or, alternatively, with predictors of a cardioprotective state, state, or cardioprotective of a predictors with alternatively, or, infarcts large developing for populations risk at of Identification knowledge gaps. high priority remain protective agent toa responsiveness clinical and procedural risk factors, different racial susceptibility profiles, and results from preclinical animal models. Indeed, models. Indeed, animal preclinical from and profiles, results different racial susceptibility factors, risk procedural and clinical the such as MI, postoperative of pathophysiology implicated inthe pathways biological in found been variability has genetic extensive that suggests state pro inflammatory of a heritability for evidence Mounting surgery. tocardiac response inflammatory acute complex injury in cardiac surgical populations, all based on a candidate gene approach. association candidate gene a on based surgical all populations, incardiac injury INTRODUCTION INTRODUCTION
locus in multiple racial groups. inmultiple racial locus artery bypass grafting (CABG) remains as high as 19%, and is associated with increased mortality and long term morbidity. long term mortality and increased with is associated and high 19%, as as remains (CABG) bypass grafting artery basis is strongly suggested by several observations, including wide variability in incidence and severity that is poorly explained by by explained is poorly that severity and inincidence variability wide observations, including several by suggested is strongly basis powered and replicated genome wide association studies (GWAS), which primarily implicate common genetic variants at the 9p21 9p21 the at variants genetic implicate common primarily which (GWAS), studies association replicated genome wide and powered 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Given the polygenic polygenic the Given 13 15 5 Of those, 1,433 patients met eligibility requirements after applying applying after requirements eligibility met patients 1,433 those, Of 18 BMJ Open http://bmjopen.bmj.com/ genome wide association study combined with a pathway analysis approach. analysis approach. pathway a with combined study association genome wide 17 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We performed joint two stage a performed We 16 The stage I cohort (discovery cohort) was comprised of 1,493 ethnically diverse subjects who underwent an isolated CABG CABG isolated an underwent who subjects diverse ethnically 1,493 of comprised was cohort) (discovery cohort stage I The Recently, integrated testing of genes involved in the same biological pathway has emerged as an alternative strategy for strategy alternative as an has emerged pathway biological inthe same involved of testing genes integrated Recently,
nature of disease susceptibility, this approach is increasingly being used to identify groups of gene variants with shared cellular shared with variants of toidentify gene groups used being increasingly is this approach susceptibility, disease of nature by this strategy, we adopted this study GWAS. of power In statistical the improving also while disease, enrichedfor are that function after MI perioperative with associated variants common genetic toidentify methodology association genome wide employing first quality control criteria and excluding patients with missing genotypes or phenotypic information. phenotypic or genotypes missing with patients criteria andexcluding control quality with cardiopulmonary bypass for the first time, and were enrolled between 2002 and 2003 in the Institutional Review Board (IRB) Board Review inthe 2003 Institutional and between 2002 enrolled and were time, first the bypass for cardiopulmonary with Ex Project of of multicenter the sub study ancillary (GeneMAGIC) Failure and Outcomes Graft Myocardial of Genetics approved,
populations Patient CABG, followed by pathway based analyses to uncover biological mechanisms of relevance to perioperative MI. toperioperative mechanisms relevance biological of touncover pathway based analyses by followed CABG, vivo Vein Graft Engineering via Transfection (PREVENT IV). Transfection via Engineering Graft Vein vivo METHODS METHODS Association Genetic of the Reporting “Strengthening by recommendations the follows the of results reporting design and study The (STREGA). Studies” evaluating the combined effects of multiple genetic variants with small effect size on a disease phenotype. disease a size effect small on with variants multiplegenetic of effects combined the evaluating Page 5 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 6 of 52 see as an elevation in plasma level of creatine kinase kinase creatine of level plasma in elevation an as 19 method to control for population stratification ( population stratification for tocontrol method 21
6 11 12 BMJ Open http://bmjopen.bmj.com/
18 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only We used principal components derived from EIGENSTRAT from derived components principal used We 20 ancestry were added to the discovery dataset leading to a total of 2,055 patients. The additional patients underwent CABG with with CABG patientsunderwent additional patients. The 2,055 of total toa leading dataset tothe discovery were added ancestry an (PEGASUS), IRB Outcomes Study Safety and Genetics Perioperative the part of as 1997 2006 bypass between cardiopulmonary Center. Medical University Duke at study longitudinal approved on the Illumina Human610 Quad BeadChip at the Duke Genomic Analysis Facility. Sample and genotype quality control of data flow of data flow control quality and genotype Sample Analysis Facility. Genomic the at Duke BeadChip Human610 Quad the on Illumina previously as equilibrium, Hardy Weinberg and SNP missingness, relatedness, cryptic check, call gender rates, of assessment included described. Supplemental Methods ). Supplemental MB isoenzyme (CK MB) that was > 10 times the upper limit of normal, as measured by a core laboratory within 24 hours after after hours 24 within laboratory by core a as measured limitnormal, of the times 10 upper > was that (CK MB) isoenzyme MB European self reported of patients 622 which additional dataset, an in expanded variants the genetic top tested stage we II, In infarction myocardial perioperative Definitionof MI of definition universal tothe defined MI according was Perioperative controls and quality Genotyping performed was cohorts both in Genotyping procedures. standard using or blood saliva whole was isolated from DNA Genomic surgery, and that was not attributable to an intervening clinical event or pre operative myocardial infarction (adjudicated by the by (adjudicated infarction myocardial pre operative or event clinical intervening toan not that and was attributable surgery, Events Clinical Committee). PREVENT IV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from arbitrarily arbitrarily quartiles of the CK MB CK MB the of quartiles CK MB quartile, and “1” for for “1” and quartile, CK MB th st and 4 and st 7 a priori a was analyses for stage I significance Statistical ). BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. see Supplemental TableS1 Supplemental see For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only CK MB quartile from the stage II cohort. Logistic regression with the same set of covariates was performed. performed. was set covariates of same the with regression stage Logistic cohort. the II from quartile CK MB th defined replication dataset to obviate type I error. In stage II analyses, the same clinical covariate and principal component principal and clinical covariate the same stage analyses, II In type error. toobviate I dataset replication defined Univariate regression analysis was performed to test differences in demographic, clinical and procedural characteristics characteristics procedural and clinical indemographic, totest differences performed analysis was regression Univariate a priori a defined as a 2 tailed p < 0.001, to balance between the overly conservative Bonferroni correction and type II error, given that we had had that we error, given II andtype correction Bonferroni conservative overly the between balance to < p 0.001, 2 tailed as a defined an adjust for population stratification ( stratification population for adjust adjustments (re estimated for the expanded dataset) were applied, with statistical significance defined as a Bonferroni adjusted p value adjustedp value Bonferroni a defined as significance statistical with applied, dataset) were expanded the for (re estimated adjustments PLINK 1.07, assuming an additive genetic model, and including significant clinical covariates and top 10 principal components to components top principal 10 and covariates significantclinical including model, and genetic 1.07, an additive assuming PLINK of 0.05/number of SNPs identified in stage I, to control the overall type I error rate. Haplotype association analysis was performed in was performed association analysis Haplotype rate. type error I overall tocontrolthe instage I, SNPs identified of 0.05/number of subjects within the 4 the within subjects Genome wide association analyses performed in the stage I cohort used multivariable logistic regression models implemented in implemented models regression logistic multivariable used inthe cohort I stage performed analyses association Genome wide distribution. For CK MB, a linear regression model was applied with adjustment for the same set of covariates as in the primary primary inthe as of set covariates same the adjustment with for applied model was regression linear a CK MB, For distribution. 1 the within subjects for “0” trait as status we re assigned subset, CK MB the extreme For analysis. analysis Statistical
the stage II cohort for genes tagged by the significant SNPs, adjusting for the same clinical covariates and principal components. components. principal and clinical covariates the same for SNPs, adjusting significant the by tagged for genes cohort stage II the CK MB of the phenotype continuous using cohort, stage II SNPs top in the the identified for were performed analyses Secondary 1 subjects the within whichincluded phenotype, CK MB dichotomized a extreme and values between patients with and without postoperative MI, and when statistically significant were used to adjust genetic association tests. tests. association used to genetic were adjust significant statistically when and MI, postoperative without patients and with between Page 7 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 8 of 52 ), 521 521 ), Figure 1 Figure ). ). ), 534,390 markers were were markers 534,390 ), ), were in high linkage inhigh were ), for 521 SNPs). The top 2 SNPs SNPs 2 top The SNPs). 521 for 5 (OR 0.46; 95% CI, 0.33 0.63), 0.33 0.63), CI, 95% 0.46; (OR 6 PAPPA2 Supplemental TableS2 Supplemental , rs2044061 on chromosome 8), and were and were 8), chromosome on rs2044061 ,
6 ). ). 8 BMJ Open ). A quantile quantile (Q Q) plot of observed vs expected p values expected vs observed plotof (Q Q) quantile quantile A ). http://bmjopen.bmj.com/ Supplemental Figure S2 Figure Supplemental (OR 0.46; 95% CI, 0.33 0.63) and 2.47x10 and 0.33 0.63) CI, 95% 0.46; (OR 6 ). Aortic cross clamp time, number of coronary artery grafts, and procedures in procedures addition and grafts, artery coronary of number time, cross clamp Aortic ). Supplemental Methods: A.1. Quality control of flow data Quality control A.1. SupplementalMethods: on October 1, 2021 by guest. Protected copyright. Supplemental TableS10 Supplemental For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Supplemental TableS2 Supplemental defined discovery threshold of p< 0.001 (minimum p = 2.76x10 = p (minimum 0.001 p< threshold of discovery defined a priori a After applying quality control criteria ( control criteria quality applying After In stage II analyses, 8 of the 521 SNPs met the Bonferroni correction threshold (p < 9.6 x 9.6 < 10 threshold (p correction of 8 Bonferroni the SNPs met 521 stage the analyses, II In Perioperative MI was observed in 112 of 1,433 patients (7.8%) and in 225 of 2,055 patients (10.9%) in stage I and stage stage and II stage inI (10.9%) patients in225 2,055 of and patients 1,433 of (7.8%) in112 observed MI was Perioperative SNPs met the the SNPsmet analyzed for association with perioperative MI in stage I. While none of the SNPs reached genome wide significance ( significance the genome wide of SNPs reached While none MI I. instage perioperative with association for analyzed disequilibrium, with p values for stage II of 2.43x10 of stage for II p values with disequilibrium, to CABG were significantly associated with perioperative MI in stage I analyses. Stage II analyses showed that extracardiac extracardiac that Stage showed analyses II analyses. instage MI I perioperative with associated significantly were toCABG ( potential confounders for tests association genetic the toadjust models regression logistic
(rs10454444 and rs10913237), located in the pregnancy associated plasma protein A2 gene ( A2 gene plasma protein pregnancy associated inthe located rs10913237), and (rs10454444 cohorts, respectively ( respectively cohorts, arteriopathy and year of surgery also played significant roles. All clinical variables identified above were included in multivariable inmultivariable included above were identified variables clinical All roles. significant played also surgery of and arteriopathy year subsequently analyzed in stage ( instage II analyzed subsequently ( for adjusted was well substructure population that showed Finally, we employed pathway analysis to prioritize association results and provide biological interpretation, using functional ontology ontology functional using interpretation, biological and provide results association prioritize to pathway analysis employed we Finally, ). Methods see Supplemental St. Joseph, MI; (GeneGO, toolsinMetaCore analysis implemented enrichment RESULTS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
-6
-6 -6 -6 -6 -6 -6 -6 ). ). 8.89 x10 2.47 x10 2.43 x10 8.18 x10 6.80 x10 9.55 x10 5.81 x10 4.70x10
a member of of member a
, SEC24 related SEC24 related ,
Table1 0.76) ataset* - d 0.63) 0.63) 3.17) 2.66) 2.05) 2.09) 0.65)
------SEC24D, SEC24D, SEC24D (1.32 OR (95%OR CI) P value 0.62 (0.50 0.46 (0.33 0.46 (0.33 2.23 (1.57 1.98 (1.47 1.65 1.67 (1.34 0.47 (0.34
Stage II: Expanded 0.11 0.11 0.11 0.16 0.36 0.36 0.33 0.11 Cases N=225
MAF
0.19 0.19 0.06 0.09 0.28 0.48 0.25 0.18 N=1,830
-4 -4 -4 -5 -4 -4 -4 -4 ), 2 in SEC24 family, member D ( D member inSEC24family, 2 ),
2.67x10 , regulatory factor X associated protein; protein; X associated factor regulatory , 1.81 x10 4.34 x10 3.13 x10 2.59 x10 3.51 x10 3.14 x10 1.85 x10
HDAC4 9 RFXAP 0.64) 3.46) 3.34) 2.33) 0.78) 2.33) 0.64) 0.66) ------ataset*
BMJ Open (1.42 http://bmjopen.bmj.com/ OR (95%OR CI) P value Controls 0.39 (0.24 2.22 2.27 (1.54 1.73 (1.29 0.58 (0.43 1.73 (1.28 0.39 (0.24 0.41 (0.25 , thyrotropin releasing hormone degrading ectoenzyme; intergenic region is expressed expressed is region intergenic ectoenzyme; hormone degrading thyrotropin releasing ,
0.09 0.13 0.18 0.40 0.35 0.35 0.09 0.09 Stage I: Discovery d Cases N=112
TRHDE MAF
on October 1, 2021 by guest. Protected copyright. 0.18 0.06 0.09 0.29 0.48 0.25 0.18 0.18 N=1,321
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
, pregnancy associated plasma protein A2; A2; protein plasma pregnancy associated , LOC552889 SMAD9
| |
For peer review only PAPPA2 HDAC4 SEC24D SEC24D PAPPA2 PAPPA2 TRHDE RFXAP PAPPA2
176649181 240234901 119691624 119710421 74471340 37417427 176630416 176566350 , mother against decapentaplegic homolog 9; homolog decapentaplegic against mother ,
237 respectively. In addition, one intronic SNP in histone deacetylase 4 ( deacetylase 4 SNP inhistone intronic one addition, In respectively.
the cytoplasmic Coat Protein II complex), and 2 located in intergenic regions were associated with perioperative MI MI ( perioperative with associated regionswere inintergenic located 2 and complex), Protein Coat II cytoplasmic the SMAD9 Top 8 Single Nucleotide Polymorphisms Associated With Postoperative Myocardial Myocardial Infarction With Postoperative Associated Polymorphisms Nucleotide Single 8 Top , histone deacetylase 4; deacetylase 4; histone , rs10454444 rs10200850 rs4834703 rs6822035 rs2303970 rs609418 rs10913 rs10489478
1 2 4 4 12 13 1 1 Chr SNP Base pair Gene symbol Controls Chr, Chromosome; MAF, minor allele frequency (based on the discovery dataset); OR, odds ratio; 95% CI, 95% confidence interval interval 95%confidence CI, 95% ratio; odds OR, dataset); discovery the on (based frequency allele minor MAF, Chromosome; Chr, Table1. HDAC4 by two franking genes with ‘|’ in between. in between. ‘|’ with genes franking two by protein D; D; protein Page 9 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 10 of 52
*
P value 0.1 0.09 0.08 0.002 0.1 SEC24D
) ) ) ) ) 1.052 1.043 2.326 2.654 1.047 - - - - - 0.578 0.575 0.950 1.242 0.574 0.78 ( 0.77 ( 1.49 ( 1.82 ( 0.77 (
) showed the strongest association, association, strongest the showed ) -4 -5 -6 -4 -4 as CK MB an extreme phenotype 5 * 1.21x10 2.27x10 8.26 x10 7.72 x10 1.30 x10 ). Overall, the rs4834703 SNP rs4834703 in the Overall, ). (p = 8.26x10 = (p Table2
5.59 5.58 5.40 9.26 8.54 - - -
HDAC4 CK MB asCK MB a quantitative trait 10 Β coefficient P value (95% OR CI) BMJ Open http://bmjopen.bmj.com/ ). When the dichotomized extreme CK MB phenotype was studied inthe was phenotype CK MB extreme dichotomized the When ).
18 07 10 18 17 0. 0. 0. 0. 0. (rs4834703 and rs6822035, p=0.002) and rs609418 in the intergenic region intergenic rs609418in the and p=0.002) rs6822035, and (rs4834703 Table2 AF (Overall) M ) and rs10200850 in rs10200850 and ) 6 SEC24D on October 1, 2021 by guest. Protected copyright. (p = 0.0001) 0.0001) ( = (p (p = 7.72x10 = (p
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
PAPPA2 PAPPA2 HDAC4 SEC24D PAPPA2 For peer review onlyPAPPA2 (p=0.001) remained nominally significant ( significant nominally remained (p=0.001) SMAD9
and and 176649181 240234901 119691624 176630416 176566350 RFXAP When CK MB was tested as a quantitative trait, all 8 SNPs remained significantly associated with plasma levels of CK MB. CK MB. of levels plasma with associated significantly all SNPstrait, 8 remained quantitative a tested as was CK MB When
followed by the 2 SNPs 2 in the by followed However, rs4834703 in SEC24D rs4834703 However, stage II cohort, only the in only cohort, SNPs 2 located stage II
consistently showed strong association signals across postoperative MI, quantitative CK MB, and the extreme CK MB traits. CK MB the extreme and CK MB, quantitative MI, postoperative across signals association strong showed consistently between between
rs10454444 rs10200850 rs4834703 rs10913237 rs10489478
1 2 4 1 1 Chr SNP Base pair Gene symbol *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of surgery, of surgery, year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted time. cross clamp aortic and vessels, of diseased number Table2. inthe Phenotype Extreme an and as Trait Quantitative as a CK MB with Polymorphisms Nucleotide Single 8 Top the of Association Joint Analysis Dataset Dataset JointAnalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
). ).
are shown in shown are 0.001 0.002 0.12 , SEC24 related SEC24 related , . Haplotype Haplotype . P
) ) ) SEC24D SEC24D 2.034 1.903 1.047 - - - , and , 1.189 1.152 0.669 1.56 ( 1.48 ( 0.84 ( HDAC4 , ,
OR (95%OR CI) -4 -4
0.01 PAPPA2 4.81 x10 2.26 x10 , OR 1.87; 95% CI, 1.46 2.39, Table3 1.46 2.39, CI, 95% 1.87; OR , 7
, regulatory factor X associated protein; protein; X associated factor regulatory , 2.93 4.42 4.80 -
11 RFXAP BMJ Open Haplotype Frequency http://bmjopen.bmj.com/ gene and results of association tests with incidence of perioperative incidence with tests association of and results gene
, thyrotropin releasing hormone degrading ectoenzyme ectoenzyme hormone degrading thyrotropin releasing , 30 47 26 0. 0. 0. SEC24D TRHDE
on October 1, 2021 by guest. Protected copyright.
LOC552889 SMAD9
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml | |
Table 3 and Supplemental TableS3 shown Supplemental and Table3 in are analysis haplotype of results the and , , pregnancy associated plasma protein A2; A2; protein plasma pregnancy associated , SNP2 SNP2 Haplotype SEC24D TRHDE RFXAP For peer review only
PAPPA2 119710421 74471340 37417427 had the most significant association with postoperative MI (p (p MI = 5.54x10 with postoperative association most significant the had Estimated haplotype frequencies in the inthe frequencies haplotype Estimated SNP1 SNP1 The haplotype structures of genic regions surrounding the significant SNPs significant in the regionssurrounding genic structures of haplotype The , mother against decapentaplegic homolog 9; homolog decapentaplegic against mother ,
analysis performed for the LD blocks containing the significant markers showed that the A A haplotype (rs6822035, rs10518325) in rs10518325) (rs6822035, haplotype A A the that showed markers the significant containing blocks the for performed LD analysis SEC24D myocardial infarction in the stage II analysis cohort (n = 2055) = (n cohort analysis stage II inthe infarction myocardial Supplemental Figure S3 Figure Supplemental
Table3. SMAD9 rs6822035 rs2303970 rs609418 , histone deacetylase 4; deacetylase 4; histone ,
4 12 13 Chr, Chromosome; MAF, minor allele frequency (based on the expanded dataset); OR, odds ratio; 95% CI, 95% confidence interval interval 95%confidence CI, 95% ratio; odds OR, dataset); expanded the on (based frequency allele minor MAF, Chromosome; Chr, *Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of surgery, of surgery, year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted 95%confidence Ratio, (Odds regression logistic or (Β coefficient) regression linear either timeusing cross clamp aortic and vessels, of diseased number analyses. interval) HDAC4 protein D; D; protein Page 11 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 12 of 52
10 10
× × 0.225 0.509 0.657 0.378 . The most significant most significant The . 8.84 5.54
1.19) 1.31) 1.70) 2.90) - - - - Reference Reference Reference 0.76 (0.48 0.87 (0.58 0.86 (0.43 1.39 (0.67 ).
0.685 0.8592 0.7126 0.0528 0.2874 1.65 (1.33-2.06) 0.2358 1.87 (1.46-2.39) 0.06231 0.07845 0.02636 PMI (n=1830)PMI Patients without
12 Supplemental Figure S4 Figure Supplemental
BMJ Open http://bmjopen.bmj.com/ 0.5927 0.9248 0.6044 0.3956 0.3631 (n=225) 0.03097 0.04425 0.03516 0.009083 cohorts ( cohorts Patients with PMI 8 9
A C A G G G G ------A C A A G C G on October 1, 2021 by guest. Protected copyright.
and cardiac surgical cardiac and
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 20 For peer review only rs6822035 rs10518325 rs11098451 rs11098451 rs11098451 rs10518325 rs10518325
rs6828577 rs6822035 A-A rs6822035 rs6828577 rs6822035 rs6822035 rs6822035 rs10518325 A-A rs10518325 rs10518325 rs10518325 remodeling” WNT, andcytoskeletal TGF, remodeling: “Cytoskeleton and matrix remodeling” extracellular adhesion: “Cell were S5 Figure shown SNPs inare Supplemental the analysis stage I for maps pathway enriched 10 top The OR, odds ratio; 95% CI, 95% confidence interval; SNP, single nucleotide polymorphism; PMI, perioperative myocardial infarction. infarction. myocardial perioperative PMI, polymorphism; nucleotide single SNP, interval; 95%confidence CI, 95% ratio; odds OR, and infarction myocardial associated with SNP previously rs10116277 the including MI, incident postoperative and locus 9p21 the at variation genetic between for association no evidence found we analyses, and II stage I factor adjusted risk multivariate In non surgical both in mortality
*Adjusted for clinical characteristics including extracardiac arteriopathy, recent myocardial infarction, procedure other than CABG only, year of year only, CABG than other procedure infarction, myocardial recent arteriopathy, extracardiac including characteristics clinical for *Adjusted time. cross clamp aortic vessels, of diseased number surgery, 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from SNPs in ) ) in single 6 ) ) and 2 intergenic SEC24D (p = 2.43x10 , SEC24D PAPPA2 HDAC4 , , ) ) or as an extreme CK MB phenotype 6 PAPPA2 ) ) in haplotype analyses, and by 7
13 BMJ Open http://bmjopen.bmj.com/ is an integral component of cytoplasmic coat protein complex II (COPII) showed the most consistent strong association with all three phenotypes SEC24D SEC24D , respectively). The most significant canonical pathway maps in stage I and II comparative comparative and II instage I pathway maps canonical most significant The respectively). , 6 on October 1, 2021 by guest. Protected copyright. ). S5 ). TablesS4 and Supplemental A A (rs6822035, rs10518325, p = 5.54x10 and 7.3x10 and For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 8 For peer review only SEC24D , respectively; respectively; , 3 and 1.1x10 and 4 Among Among these 8 SNPs, rs4834703 in We We present the first report of a 2 stage GWAS involving 225 perioperative MI cases and 1,830 controls. After accounting for regions. regions. The most significant association with perioperative MI was exhibited by rs10454444 in marker marker analyses, and secondary secondary analyses when CK MB was evaluated as a quantitative trait (rs4834703, p=2.43x10 clinical clinical and procedural covariates, we identified 8 significant SNPs mapped to 3 genes ( (rs4834703 and rs6822035, p=0.002). p=0.002). rs6822035, and (rs4834703 (enrichment p value = 1.9x10 = p value (enrichment DISCUSSION evaluated evaluated in this study. The transport protein transport machinery, a key player in vesicle trafficking of secretory proteins from the apparatus endoplasmic for delivery to downstream reticulum compartments. COPII is responsible (ER) for cargo sorting and to vesicle morphogenesis, with roles the in Golgi modulating ER exit, cell surface transport, lipid secretion and cholesterol biosynthesis, and function of G protein coupled receptors. Conditions of ischemia, oxidative injury, or acute phase response result in ER stress through accumulation of misfolded proteins, analyses were “Immune response: NFAT signaling and leukocyte interaction” and “Cell adhesion: extracellular matrix remodeling” (p matrix remodeling” extracellular adhesion: “Cell interaction” and leukocyte and signaling NFAT response: “Immune were analyses 7.2x10 = Page 13 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 14 of 52 and limiting IGF 2, its 26 Supplemental Supplemental Figure S6 and with SEC24D being the only isoform implicated in 22 23 involve characteristic skeletal malformations, whereas its
14 SEC24D BMJ Open The related protein PAPPA1 cleaves IGFBP 4 and is activated and however, the role of PAPPA2 in cardiovascular biology has not been http://bmjopen.bmj.com/ However, there are no reports of human diseases associated with genetic 27 28 25 29 29 30 on October 1, 2021 by guest. Protected copyright. ), ), a member of the HDAC family, mediates changes in chromatin structure by removing acetyl Animal models of defects in HDAC4 24 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only ). . . SEC24D groups groups from the core histones, resulting in transcriptional repression. HDAC4 is highly expressed in myocardium, where it plays an of of variants in other IGF system component genes, namely IGF 2 and the IGF 1 receptor ( Supplemental TableS5 Supplemental Histone deacetylase 4 ( extracellular extracellular matrix secretion. complete complete disruption results in early embryonic lethality. variation in variation plasma Pregnancy associated protein A2 (PAPPA2) is a metalloproteinase that regulates local insulin like growth factor (IGF) released released from vulnerable atherosclerotic plaques. PAPPA1 has been extensively studied as a cardiovascular risk biomarker for both diagnosis and prognosis of acute coronary syndrome; cardioprotective cardioprotective effects following ischemia reperfusion. which which leads to activation of the unfolded protein response (UPR) signaling pathway. If the protective mechanisms activated by the UPR are insufficient, cells die by apoptosis and autophagy. However, altered expression or function of SEC24 proteins could also explain ER trapping of misfolded proteins under conditions of ER stress, inhibits both IGF 1, with myocardial implications for inflammation linked angiogenesis and repair processes, previously previously reported. Consistent with our single marker analysis findings, GWAS pathway analysis identified significant contributions bioavailability bioavailability by specifically cleaving IGF binding protein 5 (IGFBP 5). In experimental models of myocardial ischemia, IGFBP 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from by 31 TGF β plays importantTGF β plays and complex roles 33
15 ). ). Although no direct functional roles are currently attributed to this BMJ Open http://bmjopen.bmj.com/ suggesting functional roles for HDAC4 in modulating perioperative myonecrosis, and their Supplemental Supplemental Table S7 32 on October 1, 2021 by guest. Protected copyright. , , 3 ). ). For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Consistent with this single locus gene association result, the bone morphogenic protein (BMP) For peer review 34 only ) ) gene, part of the transforming growth factor beta (TGF β) signaling pathway, which is markedly activated in SMAD9 Experimental HDAC inhibition is associated with a profound reduction in ischemia induced myocardial cell death, 31 Supplemental Figure S7 Figure Supplemental ischemia reperfusion ischemia reperfusion injury. intergenic intergenic region, a query of the Regulome and Haploreg databases reports that rs609418 is located within active regulatory elements (GATA2 transcription factor binding site by ENCODE ChIP seq and altering regulatory motifs in Gfi1 and Mef2 by Position Weight Matrix, triggering triggering preconditioning effects and promoting myocardial repair. Genetic variants in HDACs are important determinants of GWAS pathway analysis (p = 5.2x10 One of the intergenic SNPs associated with perioperative MI (rs609418) is located near the mother against decapentaplegic important important role in the regulation of gene expression and apoptosis. is involved in myocardial cell cycle progression, differentiation, and in regulating post infarction inflammatory responses. In animal models, associated signaling with resolution of inflammation, through repression of cytokine and chemokine the gene synthesis, and protection against SMAD myocardial transcription factors is susceptibility susceptibility to cardiovascular diseases, homolog homolog 9 ( infarcted infarcted myocardium. Members of the TGF β superfamily transduce their signal from the membrane to the nucleus via a distinct receptors combination of transmembrane and downstream effectors – the SMAD proteins. pathway, pathway, which also transduces its signals via a SMAD9 dependent cascade, was identified as one of the top scoring pathways in our potential use in predicting individual patient responsiveness to HDAC inhibition. inhibition. toHDAC responsiveness patient individual predicting in use potential Page 15 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 16 of 52
35 This discrepancy between convincing 36
16 Many different reasons may account for this, including inadequate 8 9 BMJ Open Taken together, in subjects with established coronary artery disease such http://bmjopen.bmj.com/ 5 38 39 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml as well as meta analyses. For peer37 review only Of note, most of the genes identified through pathway enrichment analysis in this study encode targets for therapeutic 13 locus locus and risk for perioperative MI or mortality after CABG. Indeed, Indeed, our observation is consistent with other studies showing that, although genetic variants at the 9p21 locus are associated with incident coronary artery disease, they may not be associated with the actual risk of MI. as as those included in our study, any lack of association of the 9p21 locus with subsequent MI could have resulted from the presence of coronary artery disease in both carriers and non carriers of the risk variants at the 9p21 locus, which seems to primarily mediate an phenotype. atherosclerotic Strengths of our study are 1) a relatively large population of cardiac surgery patients, 2) a prospective cohort design, and 3) a approach, replication of association findings at the gene and pathway level is much easier compared to replication at the individual SNP level. sample size, variation in study design, differences in allele frequencies, or variability in the definition of perioperative MI phenotype. MI perioperative definitionof inthe or variability frequencies, inallele differences design, instudy size, variation sample associations associations of 9p21 with greater burden of CAD but not with MI in the presence of underlying CAD has been further confirmed by nested case control studies Surprisingly, we have been unable to replicate previously reported associations between common genetic variants at the 9p21 combination combination of complementary single marker and pathway based genome wide association analyses. This approach allowed us to identify genetic variants that carry only small disease risk individually, but that jointly can perioperative MI contribute susceptibility. Furthermore, an relatively application of results large from pathway based analysis may effects add structure to on interpreting genomic data and allow exploration of cellular processes that functionally underpin the observed associations. Finally, by using this 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from The possibility of rare genetic variants that drive a pronounced Supplemental Supplemental Results B.6. ) show that, on samplebased current
17 8 10 11 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only In In conclusion, we report the first genome wide association study in a cohort of patients at risk for MI after CABG surgery. Several limitations are worth mentioning. Power calculations ( limitations Several calculations Power worth mentioning. are implicated implicated in the regulation of extracellular matrix remodeling, as well as the ER to Golgi secretory pathway potentially involved in clinical clinical phenotype was not explored in this study, because only variants with minor allele frequencies > 0.05 were assessed. Also, drugs drugs that have already been developed. Thus, by improving risk assessment and identifying allele specific therapeutic responses, further investigation of loci and pathways prioritized in this cardioprotection. study could yield actionable results for enhancing perioperative size and incidence of perioperative MI, our study can detect a genotypic relative risk ~2 with 80% power (assuming a variant with Based Based on our integrated approach utilizing both primary (perioperative MI) and secondary phenotypes (quantitative CK MB, extreme CK MB), single marker analysis and pathway analysis, we identified several polymorphisms in the insulin growth factor system functional functional studies to further elucidate the potential biological effects of the SNPs identified were not feasible due to lack of plasma or our Caucasian, and therefore predominantly patients in our in enrolled study. cohortswere Finally, the GeneMAGIC availability tissue groups. ethnic toother be cannot generalized findings 10% 10% minor allele frequency and a realistic linkage disequilibrium between the tested marker and the causal locus D’ = 0.8). Thus, although ours is the largest genetic association study of perioperative MI conducted to date, it is powered to detect only variants with common relatively large effect sizes. Most published genetic association studies of perioperative MI have reported larger effect sizes compared to ambulatory populations (OR range 1.79 3.97). Page 17 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 18 of 52 The use of such animal models 40
18 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only Currently, Currently, very few existing groups can bridge from genetics to molecular biology or cell physiology and to disease or novel therapy; we propose to examine the mechanisms by which variation in the observed genes is involved in myocardial ischemia reperfusion by injury examining their ineffects functional from patients myocardium human as cardiac undergoing wellsurgery, in our previously described preclinical rodent and swine models of cardioplegic arrest and cardiopulmonary bypass. would would allow pharmacological studies targeting the identified pathways to explore the mechanism by which novel cardioprotective drugs would attenuate myocardial ischemia reperfusion injury, with the ultimate goal of developing personalized cardioprotective adaptive adaptive responses to ER stress. As other GWAS of PMI cohorts publish their results, we intend to collaborate on conducting a meta analysis for this particular phenotype. While our GWAS results are intriguing, follow up studies are needed to translate these initial findings into the biological insights that could lead to predictive and therapeutic advances in perioperative care. For instance, in the regions of confirmed associations, causal variants will occasionally beonly among those directly genotyped. Moreover, GWAS detect almost exclusively the effects of common SNPs, offering limited power to capture any rare and structural variants, such as insertions, deletions, inversions and translocations. Detailed sequencing may be necessary to further characterize the genetic variations in the PMI associated regions to enhance the identification of causal variants. Furthermore, as in most cases, the functions of genes identified and their variants, as well as the mechanisms by which they may contribute to PMI pathophysiology, are largely unknown. strategies. strategies. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from
19 BMJ Open http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. None None For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
For peer review only This work was supported, in part, by National Institutes of Health grants R01 HL075273 and R01 HL092071 (to Dr. CONTRIBUTORSHIP STATEMENT: STATEMENT: CONTRIBUTORSHIP and analysis tothe contributed YJL of manuscript. drafting the and data, of interpretation and tothe analysis contributed MDK the of data. analysis and control quality tothe and contributed YWL YL manuscript. the of revision data and critical of interpretation Podgoreanu) Podgoreanu) and by American Heart Association grants 0256342U Podgoreanu). and 9951185U (to Dr. Mathew), and 0120492U (to ETHICSAPPROVAL: Dr. Board Review Institutional School Medicine of University Duke The STATEMENT: SHARING DATA JA, MFN, PKS, DJ and JPM contributed to the acquisition of the data and critical review of the manuscript. MVP contributed to the tothe contributed MVP manuscript. the of review and critical data the acquisition of tothe JPMcontributed PKS, DJ JA,and MFN, the versionof final the approved and review manuscript critical data, the of interpretation and design, analysis and conception manuscript. submitted INTERESTS: COMPETING FUNDING: No additional data available available data additional No Page 19 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 20 of 52 (5):443 53. (5):443 53. 357 (2):483 8, 88 e1 2. e1 2. 88 (2):483 8, 139
20 BMJ Open http://bmjopen.bmj.com/ (10):1172 9. (10):1172 9. (3):334 41. (3):334 41. (13):1675 84. (13):1675 84. 124 41 117 on October 1, 2021 by guest. Protected copyright. (3):681 9. (3):681 9. (6):585 91. (6):585 91. 154 305 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (11 Suppl):S60 5. Suppl):S60 5. (11 (5830):1488 91. (5830):1488 91. 122 316 coronary artery bypass graft surgery. J Thorac Cardiovasc Surg 2010; Surg Cardiovasc J Thorac surgery. graft bypass artery coronary 2010; and copy number variants. Nat Genet 2009; Genet Nat variants. number copy and coronary artery disease. Circulation 2008; Circulation disease. artery coronary study. Atherosclerosis 2001; Atherosclerosis study. 2007; Blood Institute Workshop. Circulation 2011; Circulation Workshop. Institute Blood graft surgery. Jama 2011; Jama surgery. graft 8. Liu KY, Muehlschlegel JD, Perry TE, et al. Common genetic variants on chromosome 9p21 predict perioperative myocardial injury after after injury myocardial perioperative predict 9p21 chromosome on variants genetic Common al. et TE, Perry JD, Muehlschlegel KY, Liu 8. 9. Muehlschlegel JD, Liu KY, Perry TE, et al. Chromosome 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation Circulation surgery. graft bypass artery after coronary mortality predicts variant 9p21 Chromosome al. et TE, Perry KY, JD,Liu Muehlschlegel 9. 6. Kathiresan S, Voight BF, Purcell S, et al. Genome wide association of early onset myocardial infarction with single nucleotide polymorphisms polymorphisms nucleotide single with infarction myocardial early onset of association Genome wide al. et S, Purcell BF, Voight S, Kathiresan 6. 5. Schunkert H, Gotz A, Braund P, et al. Repeated replication and a prospective meta analysis of the association between chromosome 9p21.3 and and 9p21.3 chromosome between theassociation of meta analysis aprospective and replication Repeated al. et P, Braund A, Gotz H, Schunkert 5. 7. Pankow JS, Folsom AR, Cushman M, et al. Familial and genetic determinants of systemic markers of inflammation: the NHLBI family heart family NHLBI the inflammation: of markers systemic of determinants genetic and al.Familial etM, Cushman AR, Folsom JS, Pankow 7. 4. McPherson R, Pertsemlidis A, Kavaslar N, et al. A common allele on chromosome 9 associated with coronary heart disease. Science Science disease. heart coronary with associated 9 chromosome allele on common al.A et N, Kavaslar A, Pertsemlidis R, McPherson 4. 3. Samani NJ, Erdmann J, Hall AS, et al. Genomewide association analysis of coronary artery disease. N Engl J Med2007; Engl N disease. artery coronary of analysis association Genomewide etal. J,HallAS, Erdmann NJ, Samani 3. 2. Schwartz Longacre L, Kloner RA, Arai AE, et al. New horizons in cardioprotection: recommendations from the 2010 National Heart, Lung, and and Lung, Heart, National the 2010 from recommendations incardioprotection: horizons New al. et AE, Arai RA, Kloner L, Longacre Schwartz 2. REFERENCES: REFERENCES: bypass artery coronary following survival and elevation enzyme of myocardial Association al. V,et K, Hasselblad MJ, Mahaffey Domanski 1. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (7):776 88. (7):776 88. 31 (19):2446 54. (19):2446 54. 294 (16):2020 35. (16):2020 35. 126 (1):31 8. (1):31 8. 120 (9):1045 53. (9):1045 53. 18 (11 Suppl):I106 12. (11Suppl):I106 12. 116
21 BMJ Open http://bmjopen.bmj.com/ (11 Suppl):S143 8. Suppl):S143 8. (11 124 (1 Suppl):I275 81. (1Suppl):I275 81. on October 1, 2021 by guest. Protected copyright. 114 (2):131 51. (2):131 51. 125 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (5830):1491 3. (5830):1491 3. (5):265 72. (5):265 72. 92 316 Statement. Hum Genet 2009; Genet Hum Statement. surgery in replicated analyses. Circulation 2011; Circulation analyses. in replicated surgery 2008; 2007; cardiac surgery. Circulation 2006; Circulation surgery. cardiac graft failure following coronary artery bypass graft surgery: PREVENT IV: a randomized controlled trial. Jama 2005; Jama trial. controlled randomized a IV: PREVENT surgery: graft bypass artery coronary following failure graft infarction in whites undergoing coronary artery bypass graft surgery. Circulation 2007; Circulation surgery. graft bypass artery coronary undergoing whites in infarction 16. Little J, Higgins JP, Ioannidis JP, et al. Strengthening the reporting of genetic association studies (STREGA): an extension of the STROBE STROBE the of extension an (STREGA): studies association of genetic the reporting al.Strengthening etJP, JP,Ioannidis J, Higgins Little 16. 13. Luo L, Peng G, Zhu Y, et al. Genome wide gene and pathway analysis. Eur J Hum Genet 2010; Genet Hum J Eur analysis. pathway and gene Y, Genome wide etal. Zhu G, Peng L, Luo 13. 15. Khoriaty R, Vasievich MP, Ginsburg D. The COPII pathway and hematologic disease. Blood 2012; Blood disease. hematologic and pathway COPII The D. Ginsburg MP, Vasievich R, Khoriaty 15. 12. Lobato RL, White WD, Mathew JP, et al. Thrombomodulin gene variants are associated with increased mortality after coronary artery bypass bypass artery coronary after mortality increased with associated are variants gene Thrombomodulin etJP,al. Mathew WD, White RL, Lobato 12. 14. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by genome wide association. Genomics Genomics association. genome wide by assessed diseases common seven of analysis Pathway NJ. Schork EJ, Topol A, Torkamani 14. 20. Helgadottir A, Thorleifsson G, Manolescu A, et al. A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science Science infarction. of myocardial risk the affects 9p21 chromosome on variant common A al. et A, Manolescu G, Thorleifsson A, Helgadottir 20. 17. Skol AD, Scott LJ, Abecasis GR, et al. Optimal designs for two stage genome wide association studies. Genet Epidemiol 2007; Epidemiol Genet studies. association genome wide two stage for designs Optimal al. et GR, Abecasis LJ, Scott AD, Skol 17. 11. Podgoreanu MV, White WD, Morris RW, et al. Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after after infarction myocardial postoperative of risk and polymorphisms gene al.Inflammatory et RW, Morris WD, White MV, Podgoreanu 11. 19. Thygesen K, Alpert JS, Jaffe AS, et al. Third Universal Definition of Myocardial Infarction. Circulation 2012; Circulation Infarction. Myocardial of Definition ThirdUniversal al. et AS, Jaffe JS, K, Alpert Thygesen 19. 18. Alexander JH, Hafley G, Harrington RA, et al. Efficacy and safety of edifoligide, an E2F transcription factor decoy, for prevention of vein ofvein for prevention decoy, factor transcription E2F an of edifoligide, safety and Efficacy al. et RA, Harrington G, JH,Hafley Alexander 18. 10. Collard CD, Shernan SK, Fox AA, et al. The MBL2 'LYQA secretor' haplotype is an independent predictor of postoperative myocardial myocardial postoperative of predictor independent an is haplotype secretor' 'LYQA MBL2 The al. et AA, Fox SK, Shernan CD, Collard 10. Page 21 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 22 of 52 (2):229 37. (2):229 37. 45
22 (3):407 15. (3):407 15. BMJ Open 29 http://bmjopen.bmj.com/ :57 67. :57 67. 47 (5):708 16. (5):708 16. 29 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml (2):469 77. (2):469 77. For peer review only 33 (8):904 9. (8):904 9. 38 (3):479 92. (3):479 92. (5):1644 53. (5):1644 53. (4):e61114. (4):e61114. 191 8 81 comparison with markers of systemic inflammation, platelet activation, and myocardial necrosis. J Am Coll Cardiol 2005; Cardiol Coll JAm necrosis. myocardial and activation, platelet inflammation, systemic of markers with comparison Cardiovasc Res 1997; Res Cardiovasc microembolisation in porcine heart. Cardiovasc Res 1995; Res Cardiovasc heart. porcine in microembolisation coronary occlusions. Cardiovasc Res 1995; Res Cardiovasc occlusions. coronary 2013; 2010; Genet 2006; Genet and human diseases. Int J Biochem Cell Biol 2014; Biol Cell J Biochem Int diseases. human and 2013; 29. Heeschen C, Dimmeler S, Hamm CW, et al. Pregnancy associated plasma protein A levels in patients with acute coronary syndromes: syndromes: coronary acute with patients inlevels protein A plasma et al. CW, Pregnancy associated Hamm S, Dimmeler C, Heeschen 29. 28. Vogt AM, Htun P, Kluge A, et al. Insulin like growth factor II delays myocardial infarction in experimental coronary artery occlusion. occlusion. artery coronary inexperimental infarction myocardial delays factor II growth et al. A, Insulin like Kluge P, Htun AM, Vogt 28. 26. Kluge A, Zimmermann R, Munkel B, et al. Insulin like growth factor I is involved in inflammation linked angiogenic processes after after processes angiogenic linked in inflammation is involved I factor growth Insulin like al. et B, Munkel R, Zimmermann A, Kluge 26. 23. Ahyi AN, Quinton LJ, Jones MR, et al. Roles of STAT3 in protein secretion pathways during the acute phase response. Infect Immun Immun Infect response. acute phase the during pathways secretion inprotein of STAT3 al.Roles et MR, Jones LJ, Quinton AN, Ahyi 23. 27. Kluge A, Zimmermann R, Munkel B, et al. Insulin like growth factor II is an experimental stress inducible gene in a porcine model of brief brief of model aporcine in gene inducible stress is experimental an II factor growth Insulin like al. et B, Munkel R, Zimmermann A, Kluge 27. 25. Baines AC, Adams EJ, Zhang B, et al. Disruption of the Sec24d gene results in early embryonic lethality in the mouse. PLoS One One PLoS mouse. in the lethality embryonic early in results gene Sec24d ofthe et al. B, Disruption Zhang EJ, Adams AC, Baines 25. 22. Fox RM, Hanlon CD, Andrew DJ. The CrebA/Creb3 like transcription factors are major and direct regulators of secretory capacity. J Cell Biol Biol J Cell capacity. secretory of regulators direct and major are factors transcription CrebA/Creb3 like The DJ. Andrew CD, Hanlon RM, Fox 22. 21. Price AL, Patterson NJ, Plenge RM, et al. Principal components analysis corrects for stratification in genome wide association studies. Nat Nat studies. association ingenome wide stratification for corrects analysis components Principal al. et RM, Plenge NJ, Patterson AL, Price 21. 24. Unlu G, Levic DS, Melville DB, et al. Trafficking mechanisms of extracellular matrix macromolecules: insights from vertebrate development development vertebrate from insights macromolecules: matrix of extracellular mechanisms Trafficking al. et DB, Melville DS, Levic G, Unlu 24. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from (6):653 57 (1):15 24. (1):15 24. 98
23 (2):85 92. (2):85 92. 1 BMJ Open http://bmjopen.bmj.com/ (3):351 60. (3):351 60. 98 on October 1, 2021 by guest. Protected copyright. (18):e211 2; author reply e11 2. e11 2. reply author (18):e211 2; 109 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only (2):184 95. (2):184 95. (10):3549 60. (10):3549 60. (4):950 6. 74 22 76 from ischemia/reperfusion injury. Circ Res 2006; Res Circ injury. ischemia/reperfusion from 2007; syndromes. Circulation 2004; Circulation syndromes. allows more accurate diagnosis of periprocedural necrosis and infarction after coronary intervention. J Am Coll Cardiol 2011; Coll Cardiol J Am intervention. coronary after infarction and necrosis ofperiprocedural diagnosis accurate more allows chronic coronary artery disease. Circ Cardiovasc Genet 2008; Genet Cardiovasc Circ disease. artery coronary chronic 2008; infarction in patients with established coronary artery disease. Circulation journal : official journal of the Japanese Circulation Society Society Circulation Japanese the of journal official : journal Circulation disease. artery coronary established with patients in infarction 2012; 61. 61. 35. Lim CC, van Gaal WJ, Testa L, et al. With the "universal definition," measurement of creatine kinase myocardial band rather than troponin troponin than rather band kinase myocardial of creatine measurement definition," "universal the With al. et L, Testa WJ,Gaal van CC, Lim 35. 34. Kempf T, Eden M, Strelau J, et al. The transforming growth factor beta superfamily member growth differentiation factor 15 protects the heart the heart protects factor 15 growth differentiation member superfamily factor beta growth transforming The al. J,et Strelau M, T, Eden Kempf 34. 33. Bujak M, Frangogiannis NG. The role of TGF beta signaling in myocardial infarction and cardiac remodeling. Cardiovasc Res Res Cardiovasc remodeling. cardiac and infarction in myocardial signaling TGF beta of role The NG. Frangogiannis M, Bujak 33. 30. Conti E, Andreotti F, Zuppi C. Pregnancy associated plasma protein a as predictor of outcome in patients with suspected acute coronary coronary acute suspected with in patients outcome of predictor aas protein plasma Pregnancy associated C. Zuppi F, Andreotti E, Conti 30. 32. Backs J, Olson EN. Control of cardiac growth by histone acetylation/deacetylation. Circ Res 2006; Res Circ acetylation/deacetylation. histone by growth cardiac of Control EN. J, Olson Backs 32. versus infarction myocardial with locus 9p21 chromosome the in of variation Association JB,al. et Muhlestein JF, Carlquist BD, Horne 37. 31. Granger A, Abdullah I, Huebner F, et al. Histone deacetylase inhibition reduces myocardial ischemia reperfusion injury in mice. Faseb J Faseb in mice. injury ischemia reperfusion myocardial reduces inhibition deacetylase Histone al. et F, Huebner I, Abdullah A, Granger 31. 36. Virani SS, Brautbar A, Lee VV, et al. Chromosome 9p21 single nucleotide polymorphisms are not associated with recurrent myocardial myocardial recurrent with associated not are polymorphisms nucleotide single 9p21 al.Chromosome etVV, Lee A, Brautbar SS, Virani 36. Page 23 of 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from Page 24 of 52 (p) for all single nucleotide polymorphisms nucleotide single all for (p) 10 ), but 521 SNPs met the prespecified discovery threshold threshold discovery prespecified the SNPs met 521 but ), (21):2234 45. (21):2234 45. −8 63
, rs2044061 on chromosome 8). chromosome on rs2044061 , 6 24 <9.09×10 :51. :51. p 3 BMJ Open http://bmjopen.bmj.com/ (9):957 70. (9):957 70. 61 on October 1, 2021 by guest. Protected copyright. For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml For peer review only collaborative meta analysis. J Am Coll Cardiol 2013; Coll Cardiol J Am meta analysis. collaborative methodology paper. Journal of cardiothoracic surgery 2008; surgery cardiothoracic of Journal paper. methodology events: a systematic review and meta analysis. J Am Coll Cardiol 2014; Coll Cardiol J Am meta analysis. and review asystematic events: (SNPs). None of the SNPs reached genome wide significance ( significance SNPs the genome wide of reached None (SNPs). 38. Chan K, Patel RS, Newcombe P, et al. Association between the chromosome 9p21 locus and angiographic coronary artery disease burden: a burden: disease artery coronary angiographic 9p21and locus chromosome the between Association et P, al. Newcombe RS, K,Patel Chan 38. LEGENDS FIGURE x axis The analysis. I ininfarction Stage myocardial perioperative with association of1. plot genome-wide Manhattan Figure the –log shows chromosome), by y axis order (colored in physical the genome represents 40. de Lange F, Yoshitani K, Podgoreanu MV, et al. A novel survival model of cardioplegic arrest and cardiopulmonary bypass in rats: a in rats: bypass cardiopulmonary and arrest cardioplegic of model survival novel A al. et MV, K,Podgoreanu Yoshitani F, Lange de 40.
39. Patel RS, Asselbergs FW, Quyyumi AA, et al. Genetic variants at chromosome 9p21 and risk of first versus subsequent coronary heart disease disease heart coronary subsequent versus of first risk and 9p21 chromosome at variants et al. Genetic AA, Quyyumi FW, Asselbergs RS, Patel 39. p< 0.001 for inclusion in Stage II analyses (minimum p = 2.76x10 = p (minimum analyses in inclusion II Stage for 0.001 p< 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 Figure 1 22 224x90mm (300 x 300 DPI) 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 SUPPLEMENTAL MATERIAL 3 4 5 6 Table of Contents 7 8 PEGASUS Investigative Team Members Page 2 9 10 Supplemental Methods Page 3 11 12 13 A.1. Quality control of data flow Page 3 14 15 A.2. ControllingFor for populationpeer stratification review only Page 4 16 17 A.3. Haplotype analysis Page 6 18 19 A.4. GWAS pathway analysis Page 6 20 21 Supplemental Results Page 7 22 23 B.1. Patient characteristics Page 7 24 25 B.2. Single-locus GWAS analysis Page 8 26 27 B.3. Haplotype analysis Page 8 28 29 B.4. Chromosome 9p21 locus analysis Page 14 30 31 B.5. GWAS pathway analysis results Page 15 32 33
B.6. Power calculation Page 21 http://bmjopen.bmj.com/ 34 35 B.7. Details of the top 521 SNPs identified in stage I GWAS analyses Page 22 36 37 38 Supplemental References Page 25 39 40 41 42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 1 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Perioperative Genetics and Safety Outcomes Study (PEGASUS) Investigative Team Members 3 4 Allen AS, Davis RD, Funk B, Gaca JG, Ginsburg GS, Glower DD, Goldstein DB, Grichnik KP, Hall RL, 5 6 Hauser E, Jones R, Kertai MD, Laskowitz DT, Li YJ, Lodge AJ, Mathew JP, Milano CA, Moretti EW, 7 8 Newman MF, Phillips-Bute B, Podgoreanu MV, Smith MP, Smith PK, Stafford-Smith M, Swaminathan 9 10 M, Welsby IJ, White WD, Willard HF 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 2 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 A. SUPPLEMENTAL METHODS 3 4 A.1. Quality control of data flow 5 6 The following quality control steps were taken to make sure that genotypes were correctly called. 7 8 Infinium BeadStudio raw data analysis: Any sample that had a very low intensity or a very low 9 • 10 11 call rate using the standard Illumina cluster (< 95%) was deleted. All SNPs that had a call 12 13 frequency below 100% were then reclustered. Any sample that was below a 98% call rate after 14 15 reclusteringFor was deleted. peer Next, all SNPs review that had a call frequency only below 99% were deleted. Any 16 17 SNPs for which >1% of samples were not called or were ambiguously called were deleted. A 18 19 total of 35 samples (13 in GeneMagic and 22 in PEGASUS cohort) were deleted during this 20 21 procedure.! 22 23 • Minor allele frequency (MAF) check for data handling accuracy: This step performs a basic 24 25 check of the data accuracy on the data flow pipeline from the output of the Illumina genotyping 26 27 facility to the analytical process. We checked the MAF report from PLINK software against the 28 29 original locus report generated by the genotyping facility, and ensured that the 2 matched exactly.! 30 31 Gender specification: This step compares the observed genotypes on chromosomes X and Y to 32 • 33 34 the gender specification obtained from the phenotype database. Samples with X-chromosomal http://bmjopen.bmj.com/ 35 36 heterozygosity that was inconsistent with reported gender were individually inspected against the 37 38 original data source. A total of 11 samples could not be reconciled and were excluded at this step 39 40 (8 in GeneMagic and 3 in PEGASUS cohort)! 41 42 • Cryptic relatedness: We estimated the sharing of genetic information between cohort participants on October 1, 2021 by guest. Protected copyright. 43 44 by calculating identity by descent (IBD) using PLINK software. All pairs with DNA samples 45 46 showing ≥ 0.125 estimated proportion of alleles IBD were inspected, and one sample from each 47 48 pair was excluded from further analyses. A total of 9 samples (all in PEGASUS cohort) were 49 50 excluded at this step. ! 51 52 Genotype missing: This step assesses whether the genotype missing is skewed toward cases or 53 • 54 55 controls, which could give rise to spurious association p-values. We used PLINK software to 56 57 perform this check on the top SNPs discussed in the paper.! 58 59 3 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 • Low MAF: We removed all SNPs with MAF < 0.01 in order to control for error in the estimation 3 4 of asymptotic p-values, as alleles with such a low MAF would have no chance of approaching 5 6 significance based on our sample size and perioperative MI incidence.! 7 8 Hardy-Weinberg Equilibrium (HWE): We used PLINK software to identify any observed 9 •
10 -6 11 genotypes that deviated from HWE. All markers that significantly deviated from HWE (p < 10 ) 12 13 were excluded.! 14 15 Analyses were performedFor using peer PLINK 1.07 , SAS/Geneticsreview version 9.2only, and R version 2.15.1. 16 17 18 19 A.2. Controlling for population stratification 20 21 We used a modified EIGENSTRAT method to control for population stratification.1 This method derives 22 23 the principal components of the correlations among the gene variants, and corrects for those correlations 24 25 in the association tests. Population structure was investigated using the EigenSoft program. All 15 26 27 principal components (PCs) were computed for both datasets. Plots between pairwise PCs, particularly 28 29 PC1 vs PC2, were generated to determine whether any obvious outliers deviated from the main cluster, 30 31 and hence, had to be excluded from subsequent analyses (Supplemental Figure S1). 32 33 http://bmjopen.bmj.com/ 34 35 36 Supplemental Figure S1. Comparison of PC1 37 38 vs PC2 reveals potential population 39 40 41 stratification in the discovery (Gene-Magic)
42 on October 1, 2021 by guest. Protected copyright. 43 dataset. 44 45 46 47 48 49 50
51 52 53
54 55 56 57 58 59 4 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 In addition, we performed multivariate regression analyses using a strategy to keep all PCs up to 6 7 the last PC with p < 0.05 in each step, to determine the number of PCs to be used for correction in the 8 9 final analysis. For instance, starting with 15 PCs in the model, if PC(i) is the last PC with p < 0.05, we 10 11 included PC(1) to PC(i) in the next multivariate model, and then repeated the process until the last PC 12 13 remained nominally significant. Using this iterative multivariate analysis, we found that the top 10 PCs 14 15 were a reasonableFor set for the peer Gene-Magic dataset review (p = 0.032), and wereonly subsequently re-estimated and 16 17 18 used for stage II analysis (Supplemental Table S1). This final set of 10 PCs was then used as covariates 19 20 to adjust for ancestry, along with clinical variables, in multivariate logistic regression analysis of 21 22 perioperative MI. 23 24 25 Supplemental Table S1: Principal component analysis of the study populations 26 Supplemental Table S1. Principal Component Analysis of the Study Populations
27 ANALYSIS OF MAXIMUM LIKELIHOOD ESTIMATES 28 29 DISCOVERY DATSET REPLICATION DATASET 30 Parameter DF Estimate Standard Wald chi- P-value DF Estimate Standard Wald chi- P-value 31 error square test error square test 32 33 Intercept 1 -2.635 0.127 434.325 <0.0001 1 -2.153 0.076 798.277 <0.0001 http://bmjopen.bmj.com/ 34 Principal component 1 1 -31.474 14.214 4.903 0.027 1 18.429 8.413 4.798 0.029 35 Principal component 2 1 -5.486 3.826 2.056 0.152 1 -4.009 3.338 1.443 0.230 36 37 Principal component 3 1 -7.575 3.497 4.691 0.03 1 -3.239 3.230 1.006 0.316 38 Principal component 4 1 -0.431 3.479 0.015 0.901 1 -1.266 3.119 0.165 0.685 39 40 Principal component 5 1 18.941 9.623 3.874 0.049 1 -1.642 4.828 0.116 0.734 41 Principal component 6 1 0.519 7.944 0.004 0.948 1 -15.494 7.678 4.073 0.044 on October 1, 2021 by guest. Protected copyright. 42 Principal component 7 1 -2.231 10.001 0.049 0.824 1 12.985 6.873 3.569 0.059 43 44 Principal component 8 1 -10.793 9.043 1.424 0.233 1 2.244 4.825 0.216 0.642 45 Principal component 9 1 29.876 11.661 6.564 0.01 1 -8.120 5.366 2.289 0.130
46 Principal component 10 1 -22.775 10.593 4.623 0.032 1 7.487 4.650 2.592 0.107 47 48 DF, degrees of freedom 49 50 51 16 52 53 54 55 56 57 58 59 5 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 A.3. Haplotype analysis 5 6 For the final candidate genes prioritized based on the significance of SNP association tests, we first 7 8 identified the linkage disequilibrium (LD) blocks harboring the significant SNPs. Within each LD block, 9 10 we performed haplotype association analysis using sliding windows of 3 markers across the region. This 11 12 was conducted for the expanded dataset, by constructing multivariable logistic regression models 13 14 adjusting for the same clinical covariates and PCs used for single marker analyses. 15 For peer review only 16 17 18 19 A.4. GWAS pathway analysis 20 21 SNPs with a covariate-adjusted association p-value < 0.01 in stage I GWAS were subjected to functional 22 23 ontology enrichment analyses using tools implemented in the MetaCore software suite (GeneGO, St. 24 25 Joseph, MI). Enrichment analysis begins by assigning SNPs to genes, and matching these geneIDs with 26 27 geneIDs in functional ontologies in highly curated canonical pathway maps in MetaCore. The aggregation 28 29 of GWAS associations for the genes in tested pathways, ie, the significance of enrichment, is computed as 30 31 the p-value of a hypergeometric model.2 Canonical pathways enriched in stage I and stage II analyses 32 33
were further compared; and the probability of a random intersection between the set of genes in common http://bmjopen.bmj.com/ 34 35 in the 2 cohorts, and functional ontology entities was estimated as p-value of a hypergeometric 36 37 distribution in MetaCore. 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 6 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B. SUPPLEMENTAL RESULTS 3 4 B.1. Patient characteristics 5 6 Supplemental Table S2. Clinical characteristics of the study populations 7 Table 1. Clinical Characteristics of the Study Populations
8 STAGE I ANALYSIS (N=1433) STAGE II ANALYSIS (N=2055) 9 No PMI, PMI No PMI PMI 10 Characteristics OR (95%CI) P-value* OR (95%CI) P-value* (n=1321) (n=112) (n=1830) (n=225) 11 12 Extracardiac arteriopathy 251 (19.0) 20 (17.9) 0.93 (0.56-1.53) 0.77 754 (41.2) 128 (56.9) 1.88 (1.42-2.49) <0.0001 13 Recent myocardial infarction 256 (19.4) 23 (20.5) 1.08 (0.67-1.74) 0.77 393 (21.5) 39 (17.3) 0.77 (0.53-1.10) 0.15 14 Procedure other than CABG only 126 (9.5) 23 (20.5) 2.45 (1.50-4.02) 0.0004 153 (8.4) 51 (22.7) 3.21 (2.26-4.58) <0.0001 15 Year of surgery by yearFor peer review only 16 1997 26 (1.4) 3 (1.3) 1.0 17 1998 70 (3.8) 8 (3.6) 0.99 (0.24-4.02) 0.98 18 1999 87 (4.8) 3 (1.3) 0.30 (0.06-1.57) 0.15 19 2000 77 (4.2) 14 (6.2) 1.58 (0.42-5.92) 0.50 20 2001 49 (2.7) 12 (5.3) 2.12 (0.55-8.20) 0.28 21 2002 146 (11.1) 9 (8.0) 1.0 188 (10.3) 20 (8.9) 0.92 (0.26-3.32) 0.90 22 2003 1175 (88.9) 103 (92.0) 1.42 (0.70-2.87) 0.33 1189 (65.0) 109 (48.4) 0.80 (0.24-2.67) 0.71 23 2004 32 (1.8) 8 (3.6) 2.17 (0.52-9.0) 0.29 24 2005 80 (4.4) 36 (16.0) 3.90 (1.11-13.72) 0.03 25 2006 32 (1.8) 12 (5.3) 3.25 (0.83-12.75) 0.09 26 Number of diseased vessels 27 0 67 (5.1) 6 (5.3) 1.0 69 (3.8) 6 (2.7) 1.0 28 1 259 (19.6) 29 (25.9) 1.25 (0.50-3.14) 0.63 276 (15.0) 32 (14.2) 1.33 (0.54-3.32) 0.54 29 2 519 (39.3) 43 (38.4) 0.93 (0.38-2.26) 0.86 595 (32.5) 52 (23.1) 1.01 (0.42-2.43) 0.99 30 3 476 (36.0) 34 (30.4) 0.80 (0.32-1.97) 0.62 885 (48.4) 130 (57.8) 1.69 (0.72-3.97) 0.22 31 4 5 (0.3) 5 (2.2) 11.5 (2.58-51.24) 0.0014 32 Aortic cross clamp time, min 51.2+36.8 71.9+32.4 1.015 (1.01-1.02) <0.0001 55.0+35.2 80.5+36.7 1.019 (1.015-1.023) <0.0001 33
Continuous variables are presented as means+standard deviation, and categorical variables as percent frequencies. CABG, coronary artery bypass http://bmjopen.bmj.com/ 34 35 Descriptivesurgery; OR (95%CI), statistics univariate of odds clinical ratio (95% variables confidence are interval) presented; *Comparisons as countmade using (percent the t test or frequency) chi-square test, for as appropriate categorical 36 37 variables and mean ± SD for continuous variables. PMI, postoperative myocardial infarction; CABG, 38 39 coronary artery bypass surgery; extracardiac arteriopathy, defined as any one or more of the following: 40 41 claudication, carotid occlusion or > 50% stenosis, and/or previous or planned intervention on the 42 on October 1, 2021 by guest. Protected copyright. 43 abdominal aorta, limb arteries or carotids; OR (95%CI), univariate odds ratio (95% confidence interval). 44 45 *Comparisons made using the t-test or Wald chi-square test, as appropriate. Clinical variables were 46 47 included in the final multivariate logistic regression analysis if significant (p < 0.05) in univariate 48 49 analyses. 50 51
52 53 54 55 56 57 58 59 7 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.2. Single-locus GWAS analysis - population structure adjustment 3 4 5 6 7 8 Supplemental Figure S2. Quantile - Quantile (Q-Q) 9 10 plot representation of the GWAS association results 11 12 13 in the discovery dataset 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 Distribution of expected (under the null hypothesis) vs observed p-values for single nucleotide 26 27 polymorphisms that are associated with perioperative myocardial infarction, based on a whole genome- 28 29 wide analysis. The red diagonal line represents the line obtained if the observed distribution did not 30 31 deviate from the expected distribution. All p-values were corrected for the inflation factor (λ = 1.0073). 32 33
http://bmjopen.bmj.com/ 34 35 B.3. Haplotype analysis of genes harboring top scoring SNPs 36 37 38 Supplemental Figure S3. Linkage disequilibrium (LD) structures of [A] PAPPA2 (1q23-q25 39 40 chromosome region), [B] SEC24D (4q26 chromosome region), and [C] HDAC4 (2q37.3 chromosome 41
42 region), and displayed as pairwise correlation plots using HapMap database (CEU European ancestry) on October 1, 2021 by guest. Protected copyright. 43 44 (Haploview). Regions of LD are shaded in bright red (strong LD) and lighter for moderate or weak LD. 45 46 The physical location of the individual SNPs associated with perioperative myocardial infarction in 2- 47 48 stage analyses is circled in red. True haplotype blocks in the population are marked with black lines in the 49 50 correlation plot. 51 52 53 54 55 56 57 58 59 8 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 A. PAPPA2 gene 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 B. SEC24D gene 23 24
25 26
27 28
29 30
31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 C. HDAC4 gene 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 9 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 05 06 06 05 06 - - - - -
10 P 0.589 0.282 0.141 0.123 0.266 0.266 0.917 0.936 0.0182 0.0184 0.0345 0.0754 0.0754 0.00199 0.00198 0.00075 1.01E 4.98E 4.98E 1.45E 6.47E 11 0.000477 0.000477 0.000248 0.000256 0.000274 0.000796 12
13 14 19.5 9.56 9.56 5.58 1.16 20.8 5.55 2.17 20.8 4.47 3.16 12.2 3.16 2.38 18.8 12.2 1.24 20.3 13.4 1.24 13.4 11.4 13.2 11.3 0.292 0.0109 STAT 15 For peer review only 0.00645 16
17
18 0.5 OR 1.43 1.08 1.29 1.22 1.29 1.19 1.25 1.23 1.23 1.17 1.58 1.28 1.62 1.28 1.49 1.02 1.49 1.02 19 0.464 0.698 0.474 0.474 0.631 0.515 0.616 0.618 20
21
22 F 0.166 0.655 0.179 0.344 0.576 0.175 0.578 0.248 0.175 0.589 0.237 0.241 0.237 0.522 0.193 0.759 0.197 0.755 0.245 0.675 0.244 0.676
23 0.0786 0.0485 0.0485 0.0795 0.0801 24
25 26
27 AA AA AA AC CA AC CC GA AG GA GA AG AG GA GA GA AG GA AG GC CG GG GG GG GG GG GG
28 29
30 HAPLOTYPE 31 10
32 33 34 http://bmjopen.bmj.com/ SNP2
35 RS909941 RS909941 RS909941 RS240104 RS240104 RS240104 RS1395460 RS1395460 RS1395460 RS3795320 RS3795320 RS3795320 RS7529392 RS7529392 RS7529392 36 RS10489478 RS10489478 RS10489478 RS10798463 RS10798463 RS10798463 RS12139318 RS12139318 RS12139318 RS12565327 RS12565327 RS12565327
PAPPA2 and HDAC4 and PAPPA2
37 38 39 SNP1
40 RS909941 RS909941 RS909941 RS240104 RS240104 RS240104 RS1395460 RS1395460 RS1395460 RS3795320 RS3795320 RS3795320 41 RS10489477 RS10489477 RS10489477 RS10489478 RS10489478 RS10489478 RS10798463 RS10798463 RS10798463 RS12139318 RS12139318 RS12139318 RS12565327 RS12565327 RS12565327
42 on October 1, 2021 by guest. Protected copyright.
43
44 BP2
45 176562796 176562796 176562796 176565307 176565307 176565307 176566350 176566350 176566350 176571431 176571431 176571431 176571578 176571578 176571578 176589732 176589732 176589732 176591291 176591291 176591291 176593524 176593524 176593524 176603840 176603840 176603840 46 47
48 BP1
49 176562608 176562608 176562608 176562796 176562796 176562796 176565307 176565307 176565307 176566350 176566350 176566350 176571431 176571431 176571431 176571578 176571578 176571578 176589732 176589732 176589732 176591291 176591291 176591291 176593524 176593524 176593524 50
Results of the haplotype analyses for for analyses haplotype the of Results
51 . 3
52 53 54 Table S Table PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 55 56 57 GENESYMBOL " " " " " " " " " " " " " " " " " " " " " " " " " " "
58 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 59
60 CHR Supplemental For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 06 06 06 06 06 05 05 ------10 0.288 0.288 0.174 0.239 0.239 0.231 0.119 0.234 0.362 0.464 0.749 0.0145 0.0145 0.0238 0.0787 0.0238 0.0146 0.0806 0.0503 0.00252 2.52E 2.56E 2.65E 2.60E 2.59E 1.79E 1.58E 11 0.000264 0.000269 0.000484 0.000371 12
13 14 22.2 13.3 1.13 22.1 13.3 1.13 22.1 5.98 1.85 12.2 5.98 1.39 22.1 9.13 1.39 22.1 5.11 3.09 12.7 5.11 1.44 2.43 1.42 5.97 3.05 18.4 3.83 18.6 15 For peer review only0.831 0.537 0.102 16
17
18 1.2 0.46 1.62 1.23 0.46 1.62 1.23 1.29 1.16 1.29 1.17 1.42 1.17 1.28 1.23 1.28 1.18 1.18 1.23 1.17 2.18 2.19 19 0.461 0.665 0.461 0.461 0.625 0.832 0.721 0.804 0.962 20
21 22 0.177 0.756 0.176 0.756 0.176 0.489 0.336 0.351 0.489 0.161 0.176 0.663 0.161 0.176 0.591 0.232 0.245 0.591 0.164 0.769 0.164 0.227 0.706 0.061 0.663 0.276 0.061
23 0.0675 0.0674 0.0665 0.0615 24 25 26
27 AA AA AA AA AA AA AA AA CA GA AG AG GA GA GA AG GA AG AG GA AA GC AG AG 28 GG GG GG GG GG GG GG 29 30 31 11
32 33 34 http://bmjopen.bmj.com/
35 RS726252 RS726252 RS726252 RS791031 RS791031 RS791031 RS4507975 RS4507975 RS4507975 RS2048765 RS2048765 RS2048765 RS6737742 36 RS12133739 RS12133739 RS12133739 RS10913237 RS10913237 RS10913237 RS10798473 RS10798473 RS10798473 RS17546000 RS17546000 RS17546000 RS10454444 RS10454444 RS10454444 RS12138021 RS12138021 RS12138021
37 38 39
40 RS726252 RS726252 RS726252 RS791031 RS791031 RS791031 RS7529392 RS7529392 RS7529392 RS2048765 41 RS12133739 RS12133739 RS12133739 RS10913237 RS10913237 RS10913237 RS10798473 RS10798473 RS10798473 RS17546000 RS17546000 RS17546000 RS10454444 RS10454444 RS10454444 RS12138021 RS12138021 RS12138021 RS13414769 RS13414769 RS13414769
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 176604564 176604564 176604564 176630416 176630416 176630416 176646383 176646383 176646383 176647000 176647000 176647000 176649181 176649181 176649181 176660247 176660247 176660247 176667810 176667810 176667810 176680137 176680137 176680137 176684432 176684432 176684432 240208154 240208154 240208154 240216021 46 47 48
49 176603840 176603840 176603840 176604564 176604564 176604564 176630416 176630416 176630416 176646383 176646383 176646383 176647000 176647000 176647000 176649181 176649181 176649181 176660247 176660247 176660247 176667810 176667810 176667810 176680137 176680137 176680137 240201797 240201797 240201797 240208154 50 51
52
53 54 HDAC4 HDAC4 HDAC4 HDAC4 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 PAPPA2 55 PAPPA2 PAPPA2 PAPPA2 PAPPA2 56 57
58 " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 06 05 05 - - - 10 0.91 0.96 0.95 0.913 0.956 0.961 0.986 0.015 0.032 0.862 0.011 0.867 0.758 0.737 0.108 0.593 0.533 0.583 0.895 0.365 0.534 0.895 0.364 0.487 0.0244 0.0575 0.0689 9.58E 2.32E 2.43E 11 0.000026 12
13
14 4.6 5.06 5.92 19.6 6.47 17.7 3.61 17.9 3.31 17.8 2.58 0.012 0.112 0.286 0.388 0.301 0.819 0.387 0.823 0.482 0.0128 0.0024 0.0301 0.0282 0.0946 0.0174 0.0174 15 0.00304 0.00252 For0.00393 0.00029 peer review only 16
17
18 1 0.8 1.01 1.01 1.01 1.32 2.21 0.75 1.02 2.14 2.13 2.12 0.84 1.06 1.07 1.09 19 0.773 0.986 0.986 0.995 0.978 0.812 0.968 0.819 0.965 0.945 0.905 0.986 0.827 0.986 0.827 20
21 22 0.51 0.73 0.52 0.13 0.718 0.221 0.221 0.185 0.594 0.182 0.594 0.224 0.266 0.221 0.208 0.414 0.524 0.415 0.423 0.512 0.488 0.382 0.546 0.382 0.723
23 0.0614 0.0615 0.0645 0.0649 0.0721 0.0721 24 25 26
27 AA AA AA AC CA AG AG GA GA AG AG GA GA AG GA AG GA GA AG AG GC CG GA 28 GG GG GG GG GG GG GG GG 29 30 31 12
32 33 34 http://bmjopen.bmj.com/
35 RS908262 RS908262 RS908262 RS925738 RS925738 RS925738 RS908265 RS908265 RS908265 RS908263 RS908263 RS908263 RS6737742 RS6737742 RS6543523 RS6543523 RS6543523 RS7594398 RS7594398 RS7594398 RS1399629 RS1399629 RS1399629 36 RS11883623 RS11883623 RS11883623 RS10200850 RS10200850 RS10200850 RS10211084 RS10211084
37 38 39
40 RS908262 RS908262 RS908262 RS925738 RS925738 RS925738 RS908265 RS908265 RS908265 RS908263 RS908263 RS2048765 RS2048765 RS6737742 RS6737742 RS6737742 RS6543523 RS6543523 RS6543523 RS7594398 RS7594398 RS7594398 RS1399629 RS1399629 RS1399629 41 RS11883623 RS11883623 RS11883623 RS10200850 RS10200850 RS10200850
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 240216021 240216021 240217116 240217116 240217116 240217522 240217522 240217522 240228263 240228263 240228263 240234901 240234901 240234901 240241255 240241255 240241255 240250456 240250456 240250456 240257958 240257958 240257958 240262287 240262287 240262287 240263103 240263103 240263103 240263584 240263584 46 47 48
49 240208154 240208154 240216021 240216021 240216021 240217116 240217116 240217116 240217522 240217522 240217522 240228263 240228263 240228263 240234901 240234901 240234901 240241255 240241255 240241255 240250456 240250456 240250456 240257958 240257958 240257958 240262287 240262287 240262287 240263103 240263103 50 51 52
53 54 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 55 HDAC4 56 57
58 " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " " 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 4 5 6 7 8
9 10 0.862 0.862 0.875 11 0.0118 0.0129 0.0122 0.0238 12
13 14 6.34 6.18 6.28 5.11 15 0.0302 0.0302 For0.0248 peer review only 16
17
18 1.29 1.27 1.02 19 0.978 0.747 0.978 0.748 20
21 22 0.31 0.52 0.17 23 0.205 0.485 0.205 0.309 24 25 26
27 AA GA GA AG GA 28 GG GG 29 30 31 13
32 33 34 http://bmjopen.bmj.com/ 35 RS6706275 RS6706275 RS6706275 36 RS10211084 RS12988669 RS12988669 RS12988669
37 38 39
40 RS908263 RS6706275 RS6706275 RS6706275 41 RS10211084 RS10211084 RS10211084
42 on October 1, 2021 by guest. Protected copyright. 43 44
45 240263584 240265617 240265617 240265617 240275570 240275570 240275570 46 47 48
49 240263103 240263584 240263584 240263584 240265617 240265617 240265617 50 51 52
53 54 55 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 HDAC4 56 57
58 " " " " " " " 2 2 2 2 2 2 2 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.4. Chromosome 9p21 locus and perioperative MI 3 4 Supplemental Figure S4. Single nucleotide polymorphisms at 9p21 locus previously associated with 5 6 incident myocardial infarction3 and mortality4 after CABG. 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 The dashed horizontal line indicates the nominal significance threshold, p = 0.05. The green dots 37 38 represent single nucleotide polymorphisms (base pair) at the 9p21 locus previously identified as 39 40 predictors of perioperative MI: rs1547705 (22082375, intron); rs1333040 (22083404, intron); rs4977574 41 on October 1, 2021 by guest. Protected copyright. 42 (22098574, intron); rs944797 (22115286, intron); rs2383207 (22115959, intron); rs1537375 (22116071, 43 44 intron). The red dot represents rs10116277 (22081397) variant, which was previously associated with 45 46 47 myocardial infarction in both non-surgical and cardiac surgical cohorts. The dark blue point rs1381923 48 49 (22909438, unknown function) is the most significant variant in this plotted region. All p-values are the 50 51 results of multivariate risk factor-adjusted analyses. 52 53 54 55 56 57 58 59 14 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.5. GWAS pathway analysis results 3 4 Of the 534,350 SNPs included in analysis, 250,740 were recognized, and 66,893 were mapped to 19,163 5 6 genes within MetaCore. Stage I cohort analysis in MetaCore identified several canonical pathways after 7 8 functional enrichment analysis with SNPs that passed the p-value and odds-ratio filters (p < 0.01, 0.9 < 9 10 OR > 1.1). 11 12 13 Supplemental Figure S5. GenGo Canonical Pathway Maps histograms after enrichment analysis with 14 15 SNPs that passedFor the threshold peer and p-value filter review in GWAS stage I analyses only. 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
http://bmjopen.bmj.com/ 34 35
36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 The top 10 canonical maps for enriched SNPs are shown as histogram plots. Results are tabulated/sorted 49 50 in the histogram based on calculated –log(p-values) of hypergeometric distribution statistics. 51 52 53 54 55 56 57 58 59 15 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Figure S6. The top scoring pathway map based on the enrichment distribution in GWAS 3 4 stage I analysis – Cell adhesion: extracellular matrix remodeling. 5 6 7 8 9 10
11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
http://bmjopen.bmj.com/ 34 35
36 37 38 39 40 41
42 Top-scoring SNPs based on GWAS stage I results are linked to genes and visualized on the maps as on October 1, 2021 by guest. Protected copyright. 43 44 thermometer-like symbols. Up-ward thermometers are red and indicate an odds ratio >1 for association 45 46 with perioperative MI, and down-ward (blue) ones indicate an odds ratio for association < 1. Letters “B”, 47 48 and “C” indicate physical interactions, B, binding; and C, cleavage. Arrows with hexagons indicate 49 50 positive (green), negative (red), and unspecified (gray) effects. Arrows without hexagons indicate 51 52 technical link (gray). For further information on pathway elements please see: 53 54 http://pathwaymaps.com/pdf/MC_legend.pdf 55 56 57 58 59 16 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Top-scoring pathways in stage I analyses were compared with enriched pathways in the stage II cohort. 3 4 The probability of a random intersection between a set of genes common to both cohorts with functional 5 6 ontology entities was estimated as p-value of hypergeometric distribution using the Compare Experiments 7 8 Workflow in MetaCore. The top 4 scored canonical pathway maps (lowest p-values), based on the 9 10 enrichment distribution, are presented below. 11 12 -4 13 1. Immune response_NFAT signaling and leukocyte interaction (p = 7.2*10 ). Nuclear factors of 14 15 activated T-cellsFor (NFATs) ispeer a family of calcium review-dependent transcription only factors with pivotal roles in 16 17 regulating immune responses, which include interactions between antigen-presenting cells and other 18 19 leukocytes, and expression of a variety of cytokines by coupling changes in intracellular calcium 20 21 concentration to gene expression. Variants in CD247, calcium channels, MHC class II, and NFAT that 22 23 map to this pathway, and their association with perioperative MI is presented below in Supplemental 24 25 Table S4. 26 27 28 SNPid Gene Symbol Gene Name OR p-value 29 30 rs2982480;rs2984800;rs2995054 CD247 CD247 molecule 2.16 0.00311 31 rs42051;rs7804449;rs1544462;rs1156325; calcium channel, voltage-dependent, alpha 2/delta 32 CACNA2D1 0.71 0.00368 rs2237526;rs2018982 subunit 1 33 34 rs1779244;rs2068357 CACNB2 calcium channel, voltage-dependent, beta 2 subunit 0.50 0.00162 http://bmjopen.bmj.com/ 35 major histocompatibility complex, class II, DP beta 2 rs3117016;rs9380342 HLA-DPB2 2.45 0.00795 36 (pseudogene) 37 rs3129882 HLA-DRA major histocompatibility complex, class II, DR alpha 1.49 0.00580 38 nuclear factor of activated T-cells, cytoplasmic, rs1562724;rs1017860;rs8090864 NFATC1 1.45 0.00362 39 calcineurin-dependent 1 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 2. Cell adhesion and extracellular matrix (ECM) remodeling (p = 1.1*10-3). An important pathway 44 45 involved in responses to myocardial injury, is comprised of matrix metalloproteinases (MMPs), their 46 47 endogenous tissue inhibitors (TIMPs), and various collagens. Furthermore, the plasminogen-plasmin 48 49 system and its regulators (PAI-1) are implicated in proteolytic degradation of ECM. Finally, 2 categories 50 51 of growth factors are part of this canonical pathway – insulin-like growth factors 1 and 2 (IGF-1 and IGF- 52 53 2), and heparin-binding EGF-like growth factor (HB-EGF) – and all are involved in regulating cell 54 55 growth, proliferation, and survival. IGF-1 and IGF-2 function by activating the IGF-1 receptor, but their 56 57 58 59 17 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 bioavailability is regulated by a family of insulin-like growth factor binding proteins (IGFBP4)., HB-EGF 3 4 activates the epidermal growth factor receptor (EGFR) and v-erb-a erythroblastic leukemia viral oncogene 5 6 homolog 4 (ErbB4), leading to cell proliferation, cell survival, and tissue remodeling (Supplemental 7 8 Table S5). 9 10
11 12 SNPid Gene Symbol Gene Name OR P-value 13 14 rs953386;rs2391823 COL4A1 collagen, type IV, alpha 1 1.76 0.00162 epidermal growth factor 15 rs884904;rs845552For peerEGFR review only1.67 0.00238 receptor 16 rs9288433;rs3817429;rs2371276;rs 17 1473636;rs7425448;rs10932384;rs6 v-erb-a erythroblastic 18 740117;rs12053361;rs17803152;rs7 ERBB4 leukemia viral oncogene 0.53 0.00613 594456;rs4672615;rs1851169;rs674 homolog 4 (avian) 19 7637 20 insulin-like growth factor 1 rs2033178 IGF1 0.50 0.00901 21 (somatomedin C) 22 rs8038056;rs8028620;rs2139924;rs insulin-like growth factor 1 IGF1R 1.90 0.00383 23 3784604 receptor 24 insulin-like growth factor 2 25 rs734351 IGF2 1.57 0.00159 26 (somatomedin A) 27 rs12454479;rs8086477;rs12454564 LAMA3 laminin, alpha 3 2.01 0.00165 28 matrix metallopeptidase 29 rs762052 MMP14 1.60 0.00863 14 (membrane-inserted) 30 31 serpin peptidase inhibitor, clade E (nexin, 32 rs1050813 SERPINE1 plasminogen activator 1.38 0.00874 33 inhibitor type 1), member
http://bmjopen.bmj.com/ 34 1 35 rs783147 PLG plasminogen 0.68 0.00979
36 37 38 3. Regulation of lipid metabolism_RXR-dependent regulation of lipid metabolism via PPAR, RAR, and
39 -3 40 VDR (p = 5.1*10 ; Supplemental Table S6). 41
42 on October 1, 2021 by guest. Protected copyright. 43 SNPid Gene Symbol Gene Name OR P-value 44 carnitine rs12364396 CPT1A 1.779 0.006433 45 palmitoyltransferase 1A 46 47 rs1286767;rs216 retinoic acid receptor, RARB 1.647 0.005116 48 4360;rs2033447 beta 49 50
51 -3 52 4. BMP (bone morphogenic protein) signaling pathway (p = 5.2*10 ). BMPs are members of the TGF-β 53 54 superfamily, with important roles in embryonic development including cardiomyogenesis and apoptosis. 55 56 BMPs transduce their signals via 2 pathways: SMAD-dependent (including SMAD9[8]), and SMAD- 57 58 59 18 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 independent cascade. BMP activity can be regulated by intracellular inhibitory SMADs (like SMAD6) or 3 4 extracellularly by secreted antagonists (like gremlin). Variants in SMAD9 and SMAD6 were associated 5 6 with an increased incidence of perioperative MI (Supplemental Table S7). 7 8 9 SNPid Gene Symbol Gene Name OR p-value 10 11 12 rs11858577 SMAD6 SMAD, mothers against DPP homolog 6 2.30 0.0002506 13 Mothers against decapentaplegic, drosophila, rs7998663 SMAD9 2.47 0.000167 14 homolog of, 9 15 rs884940For SKIpeerski oncoprotein review only0.61 0.0006749 16 17 18 19 In summary, functional analysis of genome-wide association data using biological ontologies allowed us 20 21 to supplement our single marker analysis results by identifying a number of additional candidates for 22 23 further investigation as biomarkers for perioperative MI. It also highlighted the potential mechanistic 24 25 importance of several signaling pathways that have not previously been implicated in the complex and 26 27 multifactorial pathogenesis of perioperative ischemia-reperfusion injury. 28 29 30 31 32 Supplemental Figure S7 summarizes RegulomeDB (regulome.stanford.edu) and HaploReg v2 33 34 (http://www.broadinstitute.org/mammals/haploreg/haploreg.php) database results for rs609408 (intergenic http://bmjopen.bmj.com/ 35 36 variant, located 1.5kb 3’ of SMAD9). 37 38 A. RegulomeDB supporting data for rs609418 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50
51 52
53 54 55 56 57 58 59 19 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B. HaploReg v2 summary for rs609418 and variants with r2≥0.8 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 and detailed view for rs609418 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46
47 48 49 50 51 52 53 54 55 56 57 58 59 20 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.6. Power calculation 3 4 We used Genetic Power Calculator (http://pngu.mgh.harvard.edu/~purcell/gpc/) to estimate the effect size 5 6 that our study sample size can detect with 80% power. As an example, we used rs4834703 in the SEC24D 7 8 gene, which has 0.1 MAF and odds ratio (OR) of 1.98 from the stage II analysis. In this power calculation, 9 10 we assumed 11% disease prevalence, 0.1 MAF for the marker, and the same study sample sizes (225 11 12 13 cases and 1830 controls). We varied the minor allele frequency of the disease locus (0.05, 0.1, 0.15) to 14 15 assess the genotypicFor relative riskpeer (GRR) that ourreview study sample size canonly detect with 80% power. We used 16 -5 17 the same significant threshold described in the paper, 9 x 10 , for the stage II analysis. We assumed an 18 19 additive model to define GRR of Aa and AA genotypes, that is, GRR_Aa = x and GRR_Aa = 2x, where x 20 21 is equivalent to the OR for the minor allele. If the marker tested was in complete linkage disequilibrium 22 23 (D’ = 1) with the disease locus, we found that our sample size has approximately 80% power to detect a 24 25 GRR = 2.6 for a rare disease locus (MAF = 0.05) and GRR = 1.8 for a more common disease locus (MAF 26 27 = 0.1 and 0.15). The effect sizes become slightly higher when D’ is assumed at 0.8 between disease locus 28 29 and marker (Supplemental Table S8). The OR for rs4834703 in the SEC24D gene in our study was 1.97, 30 31 which is consistent with the estimate here. Although somewhat controversial, current evidence supports 32 33
joint 2-stage analysis designs over replication analyses in GWAS, based on increased statistical power of http://bmjopen.bmj.com/ 34 35 joint analyses,5, 6 which formed the rationale for our study design. 36 37 38 Supplemental Table S8. GRR estimates that our dataset can detect a given disease locus MAF at 39 40 approximately 80% power when marker MAF = 0.1, sample size = 225 cases and 1830 controls, and 41
42 disease prevalence = 11%. on October 1, 2021 by guest. Protected copyright. 43 44 D' Disease locus MAF Odds Ratio Power 45 1 0.05 2.6 0.8 46 47 0.1 1.8 0.84 48 0.15 1.75 0.81 49 0.8 0.05 3.1 0.8 50 0.1 2.15 0.84 51 0.15 2 0.82 52 53 54 55 56 57 58 59 21 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 B.7. Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 8 RS2044061 19723503 INTS10 | LPL G 0.3117 2.004 1.499 2.68 2.76E-06 4 RS6839725 21870314 LOC100131836 | KCNIP4 A 0.09312 2.101 1.42 3.11 0.0002053 1 RS6660197 200703541 DDX59 | CAMSAP1L1 A 0.1266 2.368 1.63 3.438 5.99E-06 11 RS1816775 13783230 FAR1 | SPON1 A 0.3728 1.71 1.288 2.269 0.0002075 1 RS12756886 200840467 CAMSAP1L1 | GPR25 G 0.1215 2.342 1.612 3.404 8.12E-06 4 RS10516398 21746541 KCNIP4 A 0.09791 2.085 1.414 3.074 0.0002086 7 3 RS6788848 103028245 LOC100287880 | ALCAM G 0.237 1.995 1.473 2.702 8.15E-06 15 RS4288951 27751567 GABRG3 A 0.08947 2.283 1.476 3.532 0.0002089 1 RS12060387 210819848 HHAT A 0.05196 3.218 1.925 5.379 8.27E-06 3 RS9883142 7017919 GRM7 A 0.3052 1.717 1.29 2.285 0.0002102 8 1 RS12742404 200811765 CAMSAP1L1 G 0.1204 2.338 1.609 3.397 8.44E-06 3 RS12631018 116501764 LOC285194 | IGSF11 G 0.1684 1.913 1.358 2.696 0.0002105 6 RS3861449 148300251 SAMD5 | SASH1 A 0.2655 1.985 1.467 2.686 8.77E-06 7 RS17171847 41370672 C7orf10 | INHBA G 0.09312 2.121 1.425 3.157 0.0002106 9 15 RS894157 90168108 C15orf42 G 0.1012 2.333 1.604 3.394 9.42E-06 15 RS6494826 70049111 LOC145837 | C15orf50 A 0.4318 1.738 1.297 2.33 0.0002164 8 RS2235118 2027154 MYOM2 C 0.1319 2.184 1.536 3.107 1.38E-05 10 RS984668 124120559 BTBD16 | PLEKHA1 A 0.3542 1.736 1.296 2.325 0.0002167 10 1 RS2292096 200826769 CAMSAP1L1 G 0.1279 2.282 1.569 3.318 1.57E-05 17 RS10512598 73038132 ATP5H A 0.05634 2.532 1.547 4.144 0.0002202 4 RS4689485 6578487 MAN2B2 A 0.1606 2.054 1.479 2.851 1.71E-05 2 RS7594398 240250456 HDAC4 A 0.0695 2.28 1.47 3.536 0.000232 11 5 RS7721080 31973116 PDZD2 G 0.04521 3.173 1.868 5.391 1.94E-05 1 RS12032817 83177365 LPHN2 | TTLL7 G 0.3934 1.703 1.283 2.261 0.0002321 6 RS9382274 53877181 LRRC1 | C6orf142 A 0.07399 2.601 1.673 4.044 2.20E-05 5 RS17105323 146114766 PPP2R2B G 0.03205 3.205 1.723 5.96 0.0002336 12 3 RS1038517 59949140 FHIT A 0.0898 2.477 1.622 3.782 2.68E-05 12 RS11171580 55975171 OR6C4 | OR10P1 C 0.2335 1.772 1.306 2.404 0.0002349 5 RS331706 124890054 ZNF608 | GRAMD3 A 0.1478 2.128 1.495 3.027 2.71E-05 1 RS821717 232059805 DISC1 C 0.2107 0.4419 0.2859 0.683 0.0002365 13 4 RS4834703 119691624 SEC24D A 0.09379 2.271 1.544 3.342 3.13E-05 13 RS17424088 23087983 FGF9 | LOC646201 A 0.1055 2.068 1.404 3.046 0.0002365 4 RS4302525 6552201 PPP2R2C | MAN2B2 A 0.1457 2.039 1.458 2.853 3.18E-05 2 RS17027173 103057043 IL18RAP A 0.2156 1.814 1.32 2.493 0.0002383 14 13 RS9547679 37402440 RFXAP A 0.07962 2.63 1.662 4.162 3.63E-05 16 RS16956168 56249774 DKFZP434H168 | GNAO1 A 0.1738 1.887 1.345 2.649 0.0002397 1 RS12723891 217143384 ESRRG G 0.1103 2.126 1.479 3.056 4.62E-05 13 RS7332913 79933284 RBM26 G 0.4909 0.5783 0.4317 0.7748 0.0002424 15 19 RS3745752 48339300 CRXFor Gpeer 0.4011 1.844 1.373 2.476 review4.84E-05 13 RS2763810 79979679 RBM26onlyA 0.4906 0.5783 0.4317 0.7748 0.0002424 22 RS138195 46151880 ATXN10 G 0.371 1.819 1.362 2.429 5.04E-05 7 RS2700976 37400570 ELMO1 A 0.2302 1.756 1.3 2.373 0.0002431 16 6 RS2876669 148216298 SAMD5 | SASH1 A 0.2328 1.891 1.389 2.573 5.12E-05 14 RS8009218 100057672 CCDC85C G 0.4541 1.733 1.292 2.325 0.0002438 4 RS4631029 21796070 KCNIP4 | NCRNA00099 G 0.09885 2.214 1.507 3.253 5.14E-05 2 RS17015041 78416572 SNAR-H | REG3G G 0.1397 1.975 1.373 2.841 0.0002458 17 20 RS6067635 49698104 MULTIPLE_GENES:3755,100128598 A 0.2601 1.903 1.393 2.6 5.24E-05 22 RS6007696 47968450 LOC100289420 | FAM19A5 G 0.1765 1.878 1.341 2.631 0.0002459 16 RS12920403 19756059 IQCK G 0.1235 2.109 1.464 3.036 6.06E-05 9 RS10809097 1057797 DMRT2 | SMARCA2 G 0.1373 1.999 1.38 2.895 0.0002483 18 6 RS9405142 1632299 GMDS G 0.01484 5.575 2.4 12.95 6.46E-05 15 RS11858577 67066563 SMAD6 A 0.07422 2.3 1.473 3.591 0.0002506 9 RS12115781 112672877 MULTIPLE_GENES:445815,114299 C 0.08401 2.359 1.545 3.601 6.97E-05 13 RS2147568 79991226 RBM26 | NDFIP2 G 0.4906 0.5797 0.4329 0.7762 0.0002516 19 9 RS7042036 119766969 ASTN2 G 0.3168 1.81 1.351 2.425 7.13E-05 10 RS7910400 6310176 PFKFB3 | LOC399715 G 0.4224 0.5523 0.4018 0.7591 0.000254 19 RS11668505 48348363 CRX | SULT2A1 G 0.3984 1.803 1.347 2.414 7.43E-05 2 RS3828312 100184719 AFF3 A 0.2045 1.82 1.32 2.508 0.0002564 7 RS10270298 104210910 LHFPL3 A 0.04825 3.011 1.743 5.2 7.76E-05 12 RS10859508 93919441 MRPL42 | SOCS2 A 0.3552 0.5428 0.3911 0.7533 0.0002576 20 14 RS10140464 49210639 MDGA2 | RPS29 G 0.198 1.894 1.38 2.6 7.82E-05 4 RS6822035 119710421 SEC24D A 0.2955 1.734 1.29 2.329 0.0002593 6 RS9352851 81392952 BCKDHB | FAM46A A 0.3728 1.787 1.34 2.384 7.82E-05 9 RS7855908 71517082 PIP5K1B G 0.05466 2.712 1.588 4.631 0.0002598 21 7 RS10953524 106618074 PIK3CG | PRKAR2B G 0.499 0.5472 0.4056 0.7381 7.86E-05 7 RS2723520 17813677 AHR | SNX13 G 0.4521 1.719 1.285 2.299 0.0002609 21 RS928873 43293371 PRDM15 A 0.278 0.4756 0.3287 0.6881 8.05E-05 3 RS967331 61447737 FHIT | PTPRG A 0.1559 1.902 1.347 2.686 0.0002631 22 7 RS2299146 42033195 GLI3 A 0.302 1.763 1.33 2.337 8.10E-05 7 RS11763025 21920165 DNAH11 A 0.3502 1.692 1.275 2.244 0.0002649 8 RS6586870 19709775 INTS10 | LPL A 0.3934 1.8 1.344 2.412 8.11E-05 20 RS2425847 44907927 CDH22 | SLC35C2 G 0.3856 1.706 1.28 2.274 0.0002664 23 1 RS12122147 74838630 TNNI3K G 0.1004 2.181 1.48 3.214 8.14E-05 1 RS10489478 176566350 PAPPA2 A 0.1734 0.4052 0.2493 0.6586 0.0002668 9 RS457417 4855514 RCL1 C 0.1306 2.027 1.426 2.883 8.27E-05 10 RS1915613 66483216 LOC100287799 | ANXA2P3 A 0.248 0.4801 0.3235 0.7124 0.0002689 24 15 RS11638941 39481950 LOC100289563 | C15orf54 G 0.1424 2.008 1.418 2.843 8.47E-05 23 RS7051093 90319063 LOC100288789 | PABPC5 A 0.1842 0.2732 0.1359 0.5491 0.0002701 7 RS2058894 97296791 ACN9 | TAC1 A 0.0368 3.17 1.783 5.636 8.52E-05 6 RS6456672 25347733 LRRC16A C 0.01417 6.505 2.374 17.82 0.0002709 25 17 RS11077582 69911104 LOC124685 | SOX9 A 0.06309 2.456 1.568 3.845 8.62E-05 23 RS5940927 90309551 LOC100288789 | PABPC5 A 0.1831 0.2738 0.1362 0.5504 0.000277 8 RS13254568 11514144 BLK | GATA4 A 0.1732 1.978 1.406 2.783 8.91E-05 23 RS10522055 90307072 LOC100288789 | PABPC5 G 0.1831 0.2738 0.1362 0.5504 0.000277 26 20 RS2425810 44890910 CDH22 | SLC35C2 A 0.3924 1.787 1.335 2.391 9.38E-05 20 RS2425807 44881545 CDH22 | SLC35C2 G 0.4136 1.714 1.281 2.291 0.0002808 10 RS3847355 71725020 COL13A1 | H2AFY2 A 0.442 1.789 1.336 2.396 9.57E-05 8 RS4510897 72277458 EYA1 | MSC G 0.1552 1.898 1.343 2.682 0.0002813 27 5 RS10078649 179877533 GFPT2 | CNOT6 A 0.09818 2.174 1.471 3.212 9.80E-05 8 RS10808467 115320834 CSMD3 | TRPS1 A 0.4899 0.5774 0.4291 0.7768 0.0002854 21 RS4276100 43275896 LOC100288232 | PRDM15 G 0.2652 1.823 1.348 2.467 9.91E-05 22 RS5754826 34476759 LARGE | ISX A 0.2893 0.5043 0.3483 0.73 0.0002861 28 2 RS11686987 107128885 RGPD3 | LOC644604 A 0.03038 3.33 1.817 6.105 9.97E-05 9 RS752357 36744752 MELK | PAX5 G 0.3478 0.5406 0.3877 0.7538 0.0002875 16 RS8055711 7040933 A2BP1 A 0.2881 1.825 1.348 2.47 1.00E-04 1 RS10489477 176562608 PAPPA2 A 0.169 0.3777 0.2231 0.6396 0.0002909 29 2 RS4853374 78442014 SNAR-H | REG3G A 0.1353 2.076 1.436 3.002 0.0001033 9 RS11792612 17442000 CNTLN A 0.1798 1.812 1.314 2.5 0.0002932 19 RS7260368 57135195 ZNF71 G 0.4855 1.756 1.321 2.335 0.0001076 10 RS12780002 6308484 PFKFB3 | LOC399715 A 0.3342 0.5359 0.3821 0.7514 0.0002986 30 6 RS9361640 81392005 BCKDHB | FAM46A C 0.3013 1.793 1.333 2.412 0.0001118 8 RS6997580 115321807 CSMD3 | TRPS1 C 0.4882 0.5787 0.4301 0.7787 0.000305 10 RS2895524 12595147 CAMK1D A 0.3465 0.5135 0.3662 0.7201 0.000112 3 RS1284745 102324326 ZPLD1 | LOC100287880 G 0.3313 1.714 1.279 2.297 0.0003073 31 14 RS17112396 42397496 LRFN5 | FSCB A 0.01316 6.11 2.436 15.32 0.0001143 1 RS10801786 90595332 ZNF326 | BARHL2 C 0.1781 1.816 1.313 2.511 0.0003103 8 RS7839240 39746386 ADAM2 | IDO1 G 0.3998 1.763 1.322 2.353 0.0001152 13 RS609418 37417427 RFXAP | SMAD9 C 0.2551 1.731 1.284 2.333 0.0003138 32 4 RS2880099 156432368 LOC100287564 | GUCY1A3 A 0.1191 2.059 1.426 2.974 0.0001162 6 RS4708617 168389858 HGC6.3 | KIF25 A 0.1134 0.2382 0.1091 0.5198 0.0003142 2 RS4951964 31300544 GALNT14 A 0.02767 3.535 1.86 6.719 0.0001165 4 RS3108140 107334044 LOC100288276 A 0.3107 1.7 1.273 2.269 0.0003187 33 14 RS1242558 83235421 SEL1L | FLRT2 A 0.2912 1.785 1.329 2.398 0.0001186 2 RS1370631 78410008 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233 15 RS12913680 90098150 LOC283761 | C15orf42 A 0.1093 2.144 1.452 3.164 0.0001236 2 RS2901929 78428973 SNAR-H | REG3G G 0.141 1.949 1.355 2.803 0.0003233 http://bmjopen.bmj.com/ 2 RS2894595 227237862 KIAA1486 | IRS1 C 0.2895 1.77 1.322 2.37 0.0001245 2 RS2164850 78423757 SNAR-H | REG3G A 0.141 1.949 1.355 2.803 0.0003233 34 4 RS1398835 21849857 NCRNA00099 A 0.09987 2.12 1.444 3.113 0.0001262 16 RS4786590 5125365 ALG1 G 0.0135 6.207 2.291 16.81 0.0003296 9 RS7035935 112706482 PALM2 C 0.06748 2.47 1.555 3.923 0.0001278 14 RS4902165 63115447 FLJ43390 | KCNH5 G 0.1387 1.921 1.345 2.744 0.0003297 35 2 RS11686479 31309930 GALNT14 A 0.028 3.505 1.845 6.66 0.0001284 6 RS7753022 86176625 NT5E A 0.1397 1.975 1.362 2.864 0.0003344 7 RS6461168 2170759 MAD1L1 G 0.2756 1.766 1.318 2.365 0.0001368 21 RS2839401 43302079 PRDM15 | C2CD2 A 0.2402 0.4871 0.3288 0.7217 0.0003358 36 13 RS6563501 37408545 RFXAP | SMAD9 G 0.2257 1.827 1.34 2.491 0.0001385 9 RS11144511 78218231 OSTF1 | PCSK5 A 0.4501 1.71 1.275 2.293 0.0003408 12 RS11111722 104104227 STAB2 A 0.05162 2.752 1.634 4.633 0.0001398 11 RS12799462 13977738 FAR1 | SPON1 C 0.2034 0.4419 0.2825 0.6911 0.0003456 37 2 RS11685519 107122121 RGPD3 | LOC644604 A 0.03138 3.156 1.746 5.702 0.0001408 6 RS12192975 20431741 E2F3 A 0.05634 2.357 1.474 3.769 0.0003457 7 RS4346898 21951033 CDCA7L A 0.391 1.715 1.299 2.264 0.0001422 23 RS180484 150318158 HMGB3 | LOC100128688 G 0.3481 1.961 1.356 2.837 0.0003485 38 21 RS870531 43276783 LOC100288232 | PRDM15 A 0.3292 0.5185 0.3697 0.7273 0.0001424 10 RS161016 110039615 LOC100128304 | LOC645318 G 0.3381 0.5504 0.3967 0.7635 0.0003497 5 RS716866 168645723 SLIT3 G 0.4356 1.741 1.308 2.317 0.0001427 16 RS7188523 55313079 LOC100132339 | IRX6 A 0.04757 2.643 1.551 4.504 0.0003502 39 5 RS6595621 124989579 ZNF608 | GRAMD3 A 0.2375 1.858 1.35 2.556 0.0001429 12 RS2303970 74471340 TRHDE | LOC552889 A 0.4673 0.5823 0.4328 0.7833 0.0003512 1 RS1183044 209255814 PLXNA2 | LOC642587 A 0.4922 1.744 1.309 2.323 0.000146 5 RS297853 168628678 SLIT3 A 0.3367 1.683 1.265 2.239 0.0003513 40 7 RS4140805 7727101 RPA3 C 0.4214 1.747 1.309 2.33 0.0001501 4 RS1908125 134392080 PCDH10 | PABPC4L G 0.2068 1.757 1.29 2.394 0.0003531 5 RS10447192 8896731 LOC100128382 | SEMA5A A 0.4069 0.5487 0.4022 0.7486 0.0001528 7 RS2237429 42031224 GLI3 A 0.1883 1.791 1.301 2.466 0.0003571 41 2 RS858940 50945699 NRXN1 A 0.3836 1.743 1.307 2.324 0.0001535 8 RS6989246 2015119 MYOM2 A 0.09514 2.071 1.389 3.088 0.0003573 7 RS10263415 7724805 LOC729852 | RPA3 A 0.4244 1.755 1.312 2.349 0.000154 8 RS7820007 142731821 FLJ43860 | NCRNA00051 A 0.3921 1.695 1.269 2.264 0.0003578 42 10 RS17452064 11242471 CUGBP2 C 0.1088 2.059 1.416 2.995 0.0001571 6 RS6932311 25346376 LRRC16A A 0.01383 6.219 2.279 16.97 0.0003597 on October 1, 2021 by guest. Protected copyright. 15 RS8041127 90135196 C15orf42 A 0.1043 2.118 1.435 3.128 0.0001596 22 RS138204 46161875 ATXN10 A 0.3647 1.686 1.266 2.247 0.0003607 43 6 RS6456676 25350468 LRRC16A C 0.01181 6.973 2.543 19.12 0.0001614 12 RS2365432 74439038 TRHDE | LOC552889 A 0.4136 0.5733 0.4223 0.7783 0.0003619 7 RS3778973 2151178 MAD1L1 A 0.208 1.809 1.329 2.462 0.0001653 15 RS4777184 70007323 LOC145837 | C15orf50 A 0.365 1.704 1.271 2.284 0.0003629 44 13 RS7998663 37435150 SMAD9 G 0.07456 2.47 1.542 3.954 0.000167 7 RS10233681 71941439 CALN1 | TYW1B A 0.2554 0.4969 0.3382 0.73 0.000366 8 RS10086641 39754780 ADAM2 | IDO1 A 0.4035 1.738 1.303 2.318 0.0001693 4 RS6828577 119704501 SEC24D A 0.2939 1.713 1.274 2.302 0.0003672 45 4 RS1878943 9766888 LOC100128096 | DRD5 A 0.1926 1.848 1.342 2.546 0.0001704 7 RS17410662 96231614 FLJ42280 | SHFM1 A 0.1427 1.877 1.327 2.655 0.0003686 6 RS220702 46867717 GPR116 G 0.09885 2.161 1.446 3.231 0.0001718 6 RS744143 20418718 E2F3 A 0.07596 2.156 1.412 3.29 0.0003701 46 11 RS554341 128685761 FLI1 | KCNJ1 G 0.1334 1.946 1.374 2.754 0.0001746 20 RS6033188 11585276 JAG1 | LOC728573 G 0.1353 0.3323 0.1812 0.6095 0.0003707 20 RS7508996 40396739 CHD6 | PTPRT G 0.08603 2.294 1.487 3.539 0.000175 5 RS4298224 8894289 LOC100128382 | SEMA5A A 0.3698 0.5636 0.411 0.7729 0.0003729 47 10 RS2184035 19778072 C10orf112 A 0.04453 2.7 1.606 4.538 0.0001773 2 RS10179116 100174645 AFF3 A 0.3522 1.714 1.273 2.306 0.0003763 7 RS7802845 111723488 DOCK4 G 0.07794 2.318 1.493 3.598 0.0001797 12 RS10431428 93898858 MRPL42 | SOCS2 G 0.2831 1.769 1.292 2.422 0.0003775 12 RS1316899 55966101 OR6C4 | OR10P1 A 0.1619 1.844 1.339 2.541 0.0001813 15 RS2289486 90161355 C15orf42 G 0.09177 2.068 1.385 3.086 0.0003776 48 1 RS10454444 176649181 PAPPA2 G 0.1754 0.3867 0.2352 0.6359 0.0001814 7 RS1404508 96228434 FLJ42280 | SHFM1 A 0.1518 1.855 1.32 2.608 0.0003779 7 RS17482047 7755484 RPA3 G 0.08835 2.177 1.448 3.272 0.0001832 4 RS6854328 119683492 SEC24D A 0.2939 1.711 1.272 2.3 0.0003788 49 1 RS10913237 176630416 PAPPA2 A 0.1748 0.3866 0.2349 0.6362 0.0001846 2 RS13030183 79909989 CTNNA2 A 0.4467 1.681 1.262 2.239 0.0003823 15 RS4774964 57854873 CGNL1 | GCOM1 G 0.1859 0.4038 0.2508 0.65 0.0001892 9 RS10969620 30323457 LOC100289571 | LOC100288436 A 0.0948 1.983 1.358 2.894 0.0003898 50 12 RS3871519 93915922 MRPL42 | SOCS2 A 0.3526 0.5349 0.3851 0.7429 0.0001894 6 RS6568370 105845531 PREP C 0.1087 2.048 1.378 3.045 0.0003918 2 RS1595422 78434830 SNAR-H | REG3G A 0.1326 2.03 1.4 2.943 0.0001897 2 RS12469727 227301483 KIAA1486 | IRS1 A 0.1653 1.819 1.307 2.532 0.0003923 51 4 RS298988 119648399 SEC24D A 0.226 1.793 1.32 2.437 0.0001899 2 RS13422448 49743324 FSHR | NRXN1 C 0.3188 0.5325 0.3758 0.7545 0.0003934 4 RS1491152 130547852 C4orf33 | LOC100132483 G 0.3347 1.74 1.3 2.329 0.0001949 23 RS5941802 90288180 LOC100288789 | PABPC5 C 0.179 0.2835 0.1412 0.5693 0.000394 52 20 RS6022863 52556423 SUMO1P1 | BCAS1 A 0.4028 0.5521 0.4038 0.7548 0.0001965 16 RS899236 56352905 GNAO1 A 0.2095 1.789 1.297 2.469 0.0003944 8 RS4875192 3187176 CSMD1 A 0.03914 2.825 1.634 4.883 0.0001996 17 RS767234 69886832 LOC124685 | SOX9 A 0.2186 1.766 1.289 2.42 0.0003964 53 15 RS17738087 26905021 GABRB3 G 0.0975 2.117 1.425 3.145 0.0002022 20 RS3091869 45425918 LOC100127904 | EYA2 A 0.4325 1.665 1.256 2.207 0.0003971 3 RS17061897 60020954 FHIT A 0.0587 2.476 1.534 3.994 0.0002039 16 RS17720036 27139728 C16orf82 | G 0.1309 1.929 1.341 2.775 0.0003997 54 1 RS6577559 9238095 GPR157 | LOC727721 A 0.1231 1.985 1.382 2.85 0.000205 10 RS6593445 45359483 LOC100289168 | TMEM72 G 0.1501 1.854 1.317 2.609 0.0004009 55 56 57 58 59 22 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 2 RS1356799 140567666 LOC647012 | LRP1B C 0.1977 0.439 0.2783 0.6926 0.0004021 2 RS893745 189088249 LOC729141 G 0.4933 0.5933 0.4406 0.799 0.0005872 3 RS6770667 71873579 PROK2 | RYBP A 0.2903 1.686 1.262 2.251 0.0004067 6 RS6911476 26612363 ABT1 | ZNF322A A 0.02328 4.264 1.864 9.752 0.0005914 7 2 RS6708183 78384746 SNAR-H | REG3G A 0.3252 1.706 1.269 2.295 0.0004101 4 RS6858670 137258380 LOC100289626 | LOC646316 G 0.4656 0.598 0.4459 0.8019 0.0005918 4 RS11730104 4682267 STX18 | LOC100289434 C 0.4136 1.665 1.255 2.209 0.0004125 4 RS7688325 137260052 LOC100289626 | LOC646316 A 0.4656 0.598 0.4459 0.8019 0.0005918 8 RS1497194 115351857 CSMD3 | TRPS1 G 0.4899 0.5859 0.4355 0.7883 0.0004131 8 RS290623 115395451 CSMD3 | TRPS1 A 0.473 0.5957 0.4433 0.8006 0.0005939 8 20 RS6032661 44733811 NCOA5 | CD40 G 0.2208 0.4692 0.3083 0.7142 0.0004147 16 RS9889027 81341363 BCMO1 | GAN G 0.2362 1.703 1.257 2.308 0.0005983 20 RS6064024 52546889 SUMO1P1 | BCAS1 G 0.4265 1.682 1.26 2.244 0.0004162 16 RS8055894 21190092 TMEM159 A 0.1414 1.894 1.315 2.728 0.0006004 9 11 RS7926320 13782568 FAR1 | SPON1 A 0.388 1.664 1.254 2.207 0.0004164 7 RS1859107 21859436 DNAH11 C 0.1333 0.367 0.207 0.6506 0.0006012 15 RS8043317 27751786 GABRG3 A 0.05466 2.493 1.501 4.141 0.000418 9 RS2417753 109405667 LOC644620 G 0.1549 1.845 1.3 2.618 0.0006025 10 17 RS1126642 42989063 GFAP A 0.03552 2.741 1.565 4.8 0.0004211 19 RS296366 48374538 SULT2A1 A 0.238 1.728 1.264 2.361 0.0006055 10 RS317499 109995284 LOC100128304 | LOC645318 A 0.3357 0.5558 0.4009 0.7706 0.0004257 4 RS10516407 21883570 LOC100131836 | KCNIP4 A 0.1891 1.791 1.284 2.499 0.0006057 11 15 RS4777990 93018587 C15orf32 G 0.2817 1.715 1.27 2.316 0.0004304 10 RS3818540 97165811 SORBS1 A 0.1043 2.08 1.369 3.162 0.0006058 2 RS10495786 33559429 LTBP1 G 0.1073 1.974 1.352 2.884 0.0004312 10 RS10733970 58773025 ZWINT | IPMK A 0.02161 5.501 2.075 14.58 0.0006088 12 2 RS10200850 240234901 HDAC4 A 0.0652 2.22 1.424 3.462 0.0004344 1 RS6673391 30731688 LOC100288450 | MATN1 G 0.108 2.04 1.357 3.068 0.0006099 3 RS9868354 108840227 MORC1 | LOC100288721 G 0.02901 3.265 1.689 6.313 0.0004352 5 RS17192980 6977768 POLS | LOC442132 G 0.09959 1.951 1.331 2.859 0.0006113 13 2 RS1922351 35300334 MYADML | LOC100288911 G 0.3924 1.681 1.259 2.246 0.0004369 10 RS4750200 6308646 PFKFB3 | LOC399715 A 0.3404 0.556 0.3974 0.7779 0.0006118 16 RS2356417 71356135 FTSJD1 | CALB2 G 0.1265 1.927 1.337 2.777 0.0004381 8 RS987341 18009367 ASAH1 | NAT1 A 0.05908 2.282 1.423 3.658 0.000613 14 15 RS2028465 71659358 THSD4 C 0.2156 1.771 1.288 2.437 0.000441 16 RS7203768 73301302 LOC100288121 | LOC401859 G 0.3962 0.5831 0.4282 0.7939 0.0006145 2 RS888281 56507299 CCDC85A G 0.3252 1.662 1.252 2.208 0.0004457 18 RS9953631 39047265 LOC647946 | KC6 G 0.3239 0.5518 0.3926 0.7755 0.0006176 15 23 RS12559251 90342188 LOC100288789For | PABPC5 C 0.175peer0.2844 0.1409 0.5738 0.0004462review12 RS10770407 19087150 CAPZA3 | onlyPLEKHA5 G 0.1381 1.939 1.327 2.833 0.0006179 16 RS1424106 79216793 WWOX A 0.4686 0.5839 0.4324 0.7884 0.0004463 8 RS7819086 115272300 CSMD3 | TRPS1 A 0.427 1.645 1.237 2.188 0.0006201 16 2 RS17553946 155347161 GALNT13 | KCNJ3 A 0.2236 0.4806 0.3192 0.7236 0.000449 1 RS381115 153099345 SPRR2F | SPRR2C A 0.1171 1.965 1.335 2.892 0.0006205 10 RS12220134 97129606 SORBS1 A 0.06174 2.421 1.477 3.966 0.0004499 3 RS4927870 195820576 TFRC | FLJ25996 A 0.1191 1.975 1.338 2.917 0.0006207 17 2 RS16859880 220267737 DNPEP | LOC100288941 A 0.02126 4.044 1.853 8.828 0.0004507 12 RS4761520 93910308 MRPL42 | SOCS2 G 0.3067 0.5467 0.3869 0.7726 0.0006213 2 RS1922353 35302429 MYADML | LOC100288911 C 0.418 1.689 1.26 2.264 0.0004518 19 RS3786656 14506664 CD97 A 0.3364 1.645 1.237 2.188 0.0006225 18 21 RS2830247 27848998 CYYR1 G 0.07625 2.105 1.388 3.193 0.0004582 10 RS2171302 3815292 LOC100287172 A 0.1721 1.826 1.293 2.578 0.000624 9 RS7036825 16125443 C9orf93 | BNC2 C 0.02193 3.692 1.778 7.666 0.0004596 10 RS624964 11231831 CUGBP2 G 0.3161 1.687 1.25 2.276 0.0006241 19 20 RS2868714 44886189 CDH22 | SLC35C2 G 0.4879 1.676 1.255 2.237 0.0004617 2 RS7594456 212403434 ERBB4 A 0.3185 1.659 1.241 2.218 0.0006277 13 RS9516888 97931709 MBNL2 A 0.4308 1.645 1.245 2.174 0.0004621 8 RS3808606 39769375 ADAM2 | IDO1 A 0.4727 1.659 1.241 2.217 0.0006282 20 10 RS3932533 3925728 LOC100130652 | LOC727894 A 0.277 0.5138 0.3538 0.7461 0.000468 17 RS7215344 2430126 METT10D | PAFAH1B1 G 0.2505 0.5055 0.3418 0.7476 0.0006328 9 RS2503351 19182893 ADFP | LOC100288002 G 0.1694 1.826 1.303 2.558 0.0004689 17 RS7222013 66267691 SLC16A6 T 0.2498 1.737 1.265 2.384 0.0006334 21 19 RS10469470 17187518 MYO9B C 0.1276 0.3353 0.1817 0.6185 0.0004691 10 RS317490 110051491 LOC100128304 | LOC645318 G 0.3357 0.5664 0.4088 0.7848 0.0006345 13 RS12869943 38356247 TRPC4 G 0.2549 0.5119 0.3517 0.745 0.0004698 6 RS9463046 44802718 SUPT3H G 0.03981 2.735 1.536 4.872 0.0006349 22 2 RS328632 9247782 MBOAT2 | ASAP2 A 0.1694 0.3992 0.2386 0.6679 0.0004703 8 RS290592 115370512 CSMD3 | TRPS1 G 0.4835 0.5978 0.445 0.8031 0.0006351 7 RS2158499 88558769 C7orf62 | ZNF804B A 0.4163 0.5763 0.4231 0.7849 0.000472 11 RS1528663 13967222 FAR1 | SPON1 G 0.1916 0.4625 0.2971 0.7199 0.0006367 10 RS12766986 3335061 PITRM1 | LOC100287172 G 0.1842 1.764 1.283 2.424 0.0004726 11 RS1507526 13940539 FAR1 | SPON1 A 0.363 1.658 1.24 2.216 0.0006371 23 13 RS6491347 97930890 MBNL2 A 0.4372 1.641 1.243 2.167 0.0004732 8 RS4876325 115239083 CSMD3 | TRPS1 G 0.448 1.641 1.235 2.18 0.0006374 6 RS6910470 149552517 UST | MAP3K7IP2 A 0.05815 2.367 1.46 3.836 0.0004738 6 RS1333224 137801943 IFNGR1 | OLIG3 A 0.1134 1.968 1.334 2.902 0.0006387 24 7 RS2160138 7755797 RPA3 G 0.4565 1.676 1.255 2.239 0.0004741 15 RS8030303 61772892 RORA | VPS13C A 0.02328 3.421 1.689 6.93 0.0006395 12 RS1357241 74509618 TRHDE | LOC552889 C 0.4821 0.5933 0.4426 0.7952 0.0004769 15 RS4906904 27034085 GABRB3 | GABRA5 A 0.07119 2.198 1.399 3.455 0.0006402 25 22 RS929020 47971041 LOC100289420 | FAM19A5 A 0.17 1.836 1.306 2.583 0.0004774 1 RS563189 83284533 LPHN2 | TTLL7 A 0.4204 1.634 1.233 2.167 0.0006456 20 RS2425835 44901836 CDH22 | SLC35C2 G 0.4337 1.673 1.253 2.234 0.0004801 3 RS6437813 108734088 MORC1 A 0.03036 3.124 1.623 6.011 0.0006474 26 9 RS11143833 71433741 PIP5K1B A 0.01757 3.784 1.793 7.987 0.0004811 10 RS12570530 12547825 CAMK1D G 0.2901 0.5473 0.3871 0.7739 0.0006483 22 RS7284093 51121521 SHANK3 A 0.1492 1.876 1.318 2.671 0.000482 10 RS11591904 45346953 LOC100289168 | TMEM72 G 0.2783 1.678 1.246 2.26 0.0006578 27 18 RS2564500 57230260 CCBE1 G 0.3822 1.668 1.252 2.224 0.0004825 20 RS6127015 52540606 SUMO1P1 | BCAS1 A 0.4585 0.5941 0.4403 0.8017 0.0006598 3 RS212020 59988905 FHIT A 0.09615 2.096 1.383 3.176 0.0004827 19 RS182420 48372195 CRX | SULT2A1 G 0.2407 1.722 1.259 2.354 0.0006606 28 6 RS3869129 31410649 MICA | HCP5 A 0.2142 1.757 1.28 2.411 0.000485 10 RS6585876 86450858 LOC100131699 | GRID1 A 0.335 1.649 1.236 2.199 0.0006643 15 RS1520015 37429504 MEIS2 | TMCO5A A 0.2085 0.4596 0.2969 0.7114 0.0004863 12 RS1800159 57593894 LRP1 A 0.3188 1.646 1.235 2.193 0.0006646 29 6 RS17800315 150765510 IYD | PLEKHG1 A 0.2419 1.709 1.265 2.309 0.0004875 6 RS3101826 160584853 LOC100289162 | SLC22A2 A 0.3333 1.634 1.232 2.168 0.0006649 6 RS3899823 31410597 MICA | HCP5 A 0.2146 1.757 1.28 2.412 0.0004887 20 RS3092379 45432808 LOC100127904 | EYA2 G 0.4342 0.5947 0.4408 0.8023 0.0006705 30 6 RS11757075 150766316 IYD | PLEKHG1 A 0.2421 1.708 1.264 2.309 0.0004914 11 RS161138 110892082 LOC120364 | C11orf53 A 0.3991 0.5867 0.4314 0.7978 0.0006739 12 RS2607916 716509 NINJ2 | LOC100049716 G 0.2301 1.738 1.274 2.372 0.0004928 1 RS884940 2223866 SKI A 0.4298 1.628 1.229 2.155 0.0006749 31 15 RS8030720 90143567 C15orf42 A 0.1005 2.016 1.359 2.99 0.0004954 2 RS13392442 221100327 SLC4A3 | EPHA4 A 0.03374 2.84 1.556 5.183 0.0006756 12 RS4489787 48811100 ZNF641 | ANP32D G 0.1162 0.3285 0.1756 0.6147 0.0004976 7 RS10254402 104192398 LHFPL3 G 0.06849 2.295 1.422 3.706 0.0006758 32 14 RS2369319 96355334 TCL1A | C14orf132 A 0.1775 1.788 1.289 2.48 0.0004985 8 RS3802328 19297051 CSGALNACT1 G 0.3313 0.5617 0.4028 0.7833 0.0006762 20 RS1883838 44761896 CD40 | CDH22 A 0.4126 0.5742 0.4201 0.7848 0.0005009 5 RS256314 34617222 LOC729915 | LOC401180 A 0.03306 3.221 1.64 6.323 0.0006787 33 14 RS12897979 100056208 CCDC85C A 0.368 1.676 1.253 2.243 0.0005036 22 RS5754818 34464710 LARGE | ISX A 0.4217 0.5851 0.4295 0.7971 0.000679
15 RS6576617 27035375 GABRB3 | GABRA5 G 0.1053 2.004 1.354 2.965 0.0005056 2 RS3771166 102986222 IL18R1 A 0.3782 1.654 1.237 2.211 0.0006793 http://bmjopen.bmj.com/ 34 18 RS4797173 466351 COLEC12 A 0.2517 1.707 1.263 2.308 0.0005061 9 RS10816456 98451928 PTCH1 | LOC100287056 A 0.225 0.5024 0.3378 0.7473 0.0006798 12 RS1032725 32641200 BICD1 | FGD4 G 0.3109 1.676 1.253 2.242 0.0005065 17 RS1858990 59735821 NACA2 | BRIP1 A 0.1707 1.783 1.277 2.488 0.0006802 35 3 RS2194938 139731000 CLSTN2 A 0.1955 1.764 1.281 2.43 0.0005077 7 RS10272276 41411165 C7orf10 | INHBA G 0.198 1.75 1.267 2.417 0.000681 12 RS1694747 17617111 LOC390298 | RERGL A 0.08632 2.06 1.37 3.095 0.0005083 14 RS8017842 100055722 CCDC85C G 0.4049 1.657 1.238 2.219 0.0006851 36 3 RS12497400 59986922 FHIT G 0.1193 0.3111 0.161 0.6012 0.0005136 17 RS9910870 1471611 PITPNA | LOC100288617 A 0.2534 0.5119 0.3478 0.7535 0.0006854 18 RS4797604 1213560 LOC100130247 | C18orf2 G 0.1215 2.024 1.359 3.012 0.0005148 9 RS7873155 16118758 C9orf93 | BNC2 C 0.02868 3.443 1.686 7.033 0.0006913 37 3 RS575653 187596668 BCL6 | LOC339929 A 0.2831 1.69 1.257 2.272 0.0005167 6 RS16895923 42633368 UBR2 G 0.1616 1.794 1.28 2.515 0.000692 17 RS7502216 36612948 SOCS7 | ARHGAP23 A 0.3826 0.5652 0.4095 0.7801 0.0005191 4 RS716051 113964288 ANK2 G 0.1614 1.851 1.297 2.642 0.0006941 2 RS11888120 29322384 C2orf71 | CLIP4 A 0.2456 1.707 1.262 2.309 0.0005198 18 RS1019868 66753605 CCDC102B | DOK6 C 0.01284 19.66 3.516 109.9 0.0006952 38 2 RS2390165 166059919 SCN3A A 0.448 1.673 1.251 2.238 0.0005199 2 RS10204137 102968212 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 5 RS10041897 169893751 KCNMB1 | KCNIP1 A 0.225 0.4793 0.3164 0.7261 0.0005201 2 RS10192157 102968356 IL1RL1 A 0.3802 1.652 1.236 2.209 0.0006956 39 20 RS3092167 44922717 CDH22 | SLC35C2 A 0.4578 1.646 1.242 2.182 0.0005235 2 RS10206753 102968362 IL1RL1 G 0.3802 1.652 1.236 2.209 0.0006956 18 RS3748415 10471732 LOC100288932 / APCDD1 A 0.1144 0.2917 0.1453 0.5853 0.0005263 9 RS10901213 133247453 HMCN2 C 0.113 1.931 1.32 2.825 0.0006968 40 5 RS6891322 169897753 KCNMB1 | KCNIP1 A 0.1781 0.4204 0.2575 0.6862 0.0005278 16 RS9924119 71343397 FTSJD1 | CALB2 G 0.1184 1.915 1.315 2.788 0.0006983 7 RS17170988 37361023 ELMO1 C 0.2196 0.4803 0.3173 0.7272 0.0005287 15 RS4632084 70049424 LOC145837 | C15orf50 A 0.281 0.535 0.3726 0.7681 0.0006989 41 6 RS2223451 150767547 IYD | PLEKHG1 A 0.2422 1.701 1.259 2.296 0.0005304 6 RS9497912 148296513 SAMD5 | SASH1 A 0.1269 1.901 1.311 2.756 0.0007016
20 RS2425856 44911954 CDH22 | SLC35C2 A 0.4416 1.641 1.24 2.171 0.0005317 2 RS831010 170139346 LRP2 G 0.165 1.792 1.279 2.511 0.0007023 on October 1, 2021 by guest. Protected copyright. 42 4 RS1438217 118039852 TRAM1L1 | LOC100288955 A 0.4335 1.652 1.243 2.194 0.0005318 12 RS7314861 29238335 CCDC91 | FAR2 A 0.1987 1.777 1.274 2.479 0.0007045 12 RS10771958 32638162 BICD1 | FGD4 C 0.3107 1.667 1.248 2.226 0.0005321 16 RS4287581 7032053 A2BP1 G 0.2594 1.703 1.252 2.318 0.0007073 43 7 RS2058893 97283118 ACN9 | TAC1 A 0.113 1.882 1.316 2.691 0.0005331 10 RS17571397 12572613 CAMK1D A 0.2601 0.5233 0.3595 0.7616 0.0007182 3 RS7626013 59961159 FHIT G 0.1076 2.046 1.364 3.069 0.0005428 10 RS4021187 126756083 CTBP2 A 0.4825 0.6061 0.4535 0.8102 0.0007191 44 9 RS1972384 110210248 RAD23B | KLF4 A 0.1113 1.977 1.343 2.91 0.0005493 2 RS2048765 240208154 HDAC4 G 0.0641 2.183 1.388 3.433 0.0007214 15 RS28408948 81998391 TMC3 | MEX3B G 0.0105 7.914 2.448 25.58 0.0005494 2 RS4676312 107658016 LOC649489 | LOC729121 C 0.05162 2.39 1.442 3.961 0.0007214 45 6 RS476235 160622637 LOC100289162 | SLC22A2 A 0.3428 1.65 1.242 2.193 0.0005521 8 RS925760 115227867 CSMD3 | TRPS1 A 0.4443 1.629 1.227 2.161 0.0007231 6 RS1885487 7668923 SNRNP48 | BMP6 G 0.3222 1.71 1.261 2.318 0.0005527 15 RS6494825 70048984 LOC145837 | C15orf50 A 0.4238 0.5973 0.443 0.8053 0.0007234 46 1 RS12061205 176554721 PAPPA2 A 0.1515 0.3932 0.2315 0.6681 0.0005568 17 RS1971043 20043515 CYTSB G 0.1838 0.4333 0.2668 0.7037 0.0007244 1 RS10494201 118252257 FAM46C | LOC100131261 A 0.3151 1.658 1.244 2.209 0.0005574 7 RS11971322 41386423 C7orf10 | INHBA A 0.1866 1.76 1.268 2.442 0.0007256 47 12 RS1992136 109777361 FOXN4 | MYO1H G 0.1103 0.3186 0.1664 0.6102 0.0005604 12 RS7305820 74550782 TRHDE | LOC552889 G 0.4953 1.635 1.229 2.175 0.0007274 7 RS10486908 82860532 PCLO | SEMA3E A 0.05803 2.398 1.459 3.942 0.0005634 10 RS11188295 97125874 SORBS1 A 0.08907 2.149 1.379 3.35 0.0007291 48 2 RS6744657 155012888 GALNT13 A 0.237 0.4909 0.3276 0.7357 0.0005662 3 RS3772460 59943532 FHIT G 0.1086 2.005 1.339 3.003 0.0007339 12 RS10861433 105877556 C12orf75 | NUAK1 A 0.226 0.4811 0.3173 0.7294 0.000568 9 RS295263 4840063 RCL1 G 0.1208 1.869 1.3 2.686 0.0007348 49 3 RS9860107 60040105 FHIT G 0.07996 2.157 1.393 3.34 0.0005682 18 RS11080417 10423834 VAPA | APCDD1 G 0.2679 1.698 1.249 2.309 0.000736 9 RS11103417 137494787 RXRA | COL5A1 G 0.2436 0.5026 0.3398 0.7432 0.0005683 17 RS7220740 59740793 NACA2 | BRIP1 A 0.3133 1.641 1.231 2.188 0.0007379 50 4 RS2196716 118055429 TRAM1L1 | LOC100288955 C 0.4764 0.5955 0.4434 0.7997 0.0005697 2 RS6431565 238836683 RAMP1 | UBE2F A 0.3316 1.67 1.24 2.249 0.0007438 15 RS1435835 26412976 LOC100128714 | GABRB3 G 0.1292 1.931 1.328 2.808 0.0005763 12 RS12823898 93860565 UBE2N | MRPL42 A 0.114 1.978 1.331 2.939 0.0007443 51 21 RS4919983 43301781 PRDM15 | C2CD2 A 0.3155 0.5512 0.3926 0.7737 0.0005765 4 RS1395114 113953581 ANK2 A 0.191 1.797 1.278 2.526 0.0007508 1 RS12067235 205281255 NUAK2 A 0.2213 0.4798 0.3158 0.7291 0.0005813 11 RS627354 74028944 P4HA3 | PGM2L1 A 0.4325 1.621 1.224 2.146 0.0007512 52 8 RS6985983 115289224 CSMD3 | TRPS1 A 0.4265 1.649 1.24 2.193 0.0005824 1 RS10462021 7897133 PER3 G 0.1842 1.744 1.262 2.411 0.0007523 8 RS1552237 87045760 REXO1L2P | PSKH2 A 0.2854 1.672 1.247 2.241 0.0005854 1 RS12125785 209254017 PLXNA2 | LOC642587 A 0.4015 0.595 0.4398 0.8049 0.0007573 9 RS10739231 98453680 PTCH1 | LOC100287056 A 0.4281 0.5842 0.43 0.7936 0.0005855 13 RS9531028 80767832 LOC729479 | SPRY2 A 0.1899 0.4483 0.2811 0.7151 0.0007574 53 12 RS4767252 115187724 TBX3 | MED13L A 0.33 1.67 1.247 2.237 0.0005858 19 RS11083726 29388987 LOC148189 | LOC148145 A 0.4565 1.645 1.231 2.198 0.0007601 12 RS995603 29235933 CCDC91 | FAR2 C 0.1993 1.785 1.283 2.484 0.0005862 7 RS4730427 78285960 MAGI2 A 0.1309 0.3623 0.2006 0.6542 0.0007604 54 7 RS206196 20361873 MACC1 | ITGB8 A 0.4213 0.5846 0.4305 0.7939 0.0005865 1 RS7527871 2281726 MORN1 | LOC100129534 C 0.1715 1.769 1.269 2.465 0.0007626 55 56 57 58 59 23 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 Supplemental Table S9. Details of the top 521 SNPs identified in discovery dataset (cont) 3 4 Minor Minor allele Odds 95% CI 95% CI Minor Minor allele 95% CI 95% CI CHR SNP BP GENE SYMBOL P-value CHR SNP BP GENE SYMBOL Odds ratio P-value 5 allele frequency ratio Lower Limit Upper Limit allele frequency Lower Limit Upper Limit 6 12 RS11046362 22460338 ST8SIA1 A 0.1046 2.003 1.336 3.001 0.000764 5 RS31036 146615260 STK32A C 0.3286 1.644 1.227 2.203 0.0008802 7 RS2204779 118438184 ANKRD7 | LOC100287273 G 0.01619 4.926 1.946 12.47 0.0007653 7 RS974060 97271508 ACN9 | TAC1 G 0.2939 1.643 1.226 2.201 0.0008811 8 RS1606891 115206234 CSMD3 | TRPS1 A 0.4483 1.627 1.225 2.16 0.0007696 2 RS10183733 135071570 MGAT5 A 0.1772 1.773 1.265 2.484 0.0008825 7 12 RS12368128 125648067 AACS | TMEM132B A 0.2449 0.5034 0.3375 0.751 0.0007698 18 RS4306606 55334295 ATP8B1 A 0.2713 0.5341 0.3691 0.773 0.0008836 4 RS7656244 65170415 SRD5A2L2 A 0.1859 0.4634 0.2959 0.7255 0.0007708 18 RS2345 75279459 GALR1 | LOC100132713 G 0.06883 2.148 1.368 3.372 0.0008951 8 2 RS6547192 78434082 SNAR-H | REG3G A 0.1373 1.89 1.304 2.739 0.0007725 3 RS4974500 134109630 AMOTL2 | ANAPC13 G 0.4943 0.6023 0.4466 0.8124 0.0008961 2 RS6432964 149986664 LYPD6B G 0.3654 0.5723 0.4133 0.7924 0.0007749 3 RS1879784 183203034 MCF2L2 | KLHL6 A 0.4757 1.643 1.226 2.202 0.000898 9 8 RS167069 115423582 CSMD3 | TRPS1 G 0.4848 0.6037 0.4497 0.8102 0.0007753 4 RS12511494 113770535 ANK2 A 0.141 1.859 1.289 2.682 0.0009018 8 RS10283337 1319724 LOC401442 | DLGAP2 G 0.4052 1.625 1.224 2.157 0.0007815 6 RS2474880 63460047 KHDRBS2 | LOC100128610 A 0.4325 1.638 1.224 2.192 0.0009037 10 4 RS7676913 118043203 TRAM1L1 | LOC100288955 C 0.4706 0.6035 0.4495 0.8104 0.0007854 15 RS11636337 70044354 LOC145837 | C15orf50 G 0.251 0.5284 0.3625 0.7704 0.0009116 1 RS493631 153143048 SPRR2G | LELP1 A 0.1195 1.968 1.325 2.921 0.0007874 11 RS1852755 13996686 SPON1 G 0.2611 0.529 0.3631 0.7707 0.000912 11 14 RS878554 94606179 IFI27L2 | PPP4R4 A 0.2057 1.746 1.261 2.418 0.0007884 9 RS16933086 36215581 GNE A 0.2016 1.685 1.238 2.293 0.0009121 11 RS10894162 129781016 PRDM10 A 0.02092 3.344 1.652 6.767 0.000792 8 RS12547514 55806439 LOC100287651 | XKR4 C 0.4172 1.638 1.224 2.194 0.0009135 12 5 RS1154758 162716850 GABRG2 | CCNG1 G 0.4443 0.5973 0.4421 0.8071 0.0007932 7 RS13237260 7779091 RPA3 | LOC729852 A 0.2415 1.665 1.232 2.252 0.000918 10 RS7914288 71739556 COL13A1 | H2AFY2 G 0.389 1.629 1.225 2.167 0.0007965 13 RS9532283 39394277 FREM2 A 0.4673 1.626 1.22 2.168 0.0009189 13 17 RS7219555 59740122 NACA2 | BRIP1 C 0.3151 1.636 1.227 2.181 0.0007979 2 RS12619691 35306866 MYADML | LOC100288911 A 0.3937 1.628 1.22 2.172 0.0009197 12 RS10783915 59001589 XRCC6BP1 | LRIG3 G 0.332 0.5723 0.413 0.793 0.000798 3 RS6766938 67150565 KBTBD8 | SUCLG2 A 0.2142 1.721 1.248 2.373 0.0009197 14 5 RS951087 13328289 CTNND2 | DNAH5 A 0.1107 1.971 1.325 2.931 0.0008055 11 RS1852757 13987892 SPON1 G 0.1866 0.4711 0.3018 0.7353 0.0009207 13 RS9510573 23680990 LOC646201 | SGCG A 0.1889 1.746 1.26 2.419 0.0008068 9 RS7846713 8079286 C9orf123 | PTPRD A 0.2375 1.7 1.242 2.327 0.0009219 15 14 RS885845 96729728 BDKRB1 ForA 0.37peer1.639 1.228 2.189 0.0008088review5 RS457638 80359968 RASGRF2 onlyG 0.1867 0.468 0.2986 0.7333 0.0009227 1 RS6676869 209249614 PLXNA2 | LOC642587 G 0.3036 0.5568 0.3952 0.7844 0.0008115 19 RS9917042 14494473 CD97 G 0.3212 1.623 1.219 2.162 0.0009241 16 6 RS9283918 53926867 C6orf142 A 0.05904 2.267 1.404 3.661 0.0008121 8 RS12677088 19309945 CSGALNACT1 A 0.3126 0.5637 0.4015 0.7913 0.0009254 6 RS9296737 53926716 C6orf142 G 0.05904 2.267 1.404 3.661 0.0008121 6 RS6570859 148829305 SASH1 G 0.3333 0.5748 0.4141 0.7977 0.0009283 17 3 RS9841113 116455714 LOC285194 | IGSF11 A 0.4339 1.62 1.222 2.149 0.0008144 15 RS7183427 39503508 LOC100289563 | C15orf54 C 0.1835 1.766 1.261 2.472 0.0009291 5 RS1023617 172740329 NKX2-5 | STC2 A 0.3128 1.66 1.234 2.235 0.0008162 10 RS11003908 55645842 PCDH15 A 0.02092 4.64 1.87 11.51 0.0009301 18 3 RS4076927 46730749 ALS2CL G 0.3171 0.55 0.3875 0.7807 0.0008215 14 RS1197382 81146776 C14orf145 G 0.08974 0.2693 0.1238 0.5856 0.0009328 8 RS10091115 62684067 LOC100287835 | NKAIN3 C 0.3381 0.5715 0.4117 0.7934 0.0008288 7 RS3800895 2125942 MAD1L1 A 0.2743 1.636 1.222 2.19 0.0009334 19 12 RS2364227 93893510 MRPL42 G 0.303 0.5558 0.3939 0.7844 0.0008315 10 RS7905027 124114432 BTBD16 | PLEKHA1 G 0.3932 0.5897 0.4313 0.8063 0.0009347 21 RS2832279 30613466 C21orf109 A 0.3323 0.5744 0.415 0.7951 0.0008317 2 RS6748959 175424127 GPR155 | WIPF1 G 0.4906 0.6052 0.4495 0.8148 0.0009351 20 4 RS12508492 112696927 LOC729065 | C4orf32 G 0.05331 2.324 1.417 3.81 0.0008328 9 RS10814369 36292975 GNE | LOC646993 A 0.192 1.714 1.246 2.359 0.0009351 6 RS6940801 169655667 THBS2 | LOC100289536 G 0.1498 1.779 1.269 2.493 0.0008332 11 RS12793784 26700112 SLC5A12 A 0.1208 1.926 1.306 2.84 0.0009361 21 3 RS17485052 138638081 PIK3CB | FOXL2 A 0.1991 1.75 1.26 2.43 0.0008398 16 RS6564372 73324074 LOC100288121 | LOC401859 A 0.4986 0.6197 0.4668 0.8228 0.0009375 12 RS10746042 106111254 C12orf75 | NUAK1 G 0.2399 1.695 1.244 2.311 0.0008398 2 RS7559847 222068646 SLC4A3 | EPHA4 A 0.385 1.639 1.223 2.197 0.0009419 22 22 RS138154 46107105 ATXN10 G 0.2319 1.673 1.237 2.263 0.0008421 3 RS9790156 139814577 CLSTN2 A 0.1002 1.978 1.32 2.964 0.0009425 1 RS12120640 74821837 TNNI3K G 0.1205 1.86 1.292 2.678 0.0008424 1 RS1925421 39250964 LOC400750 | RRAGC A 0.2567 1.645 1.225 2.21 0.0009447 23 20 RS2425817 44895286 CDH22 | SLC35C2 A 0.4329 1.633 1.225 2.179 0.0008438 23 RS10521479 55583383 LOC644893 | FOXR2 A 0.1163 2.228 1.386 3.582 0.0009464 7 RS4730197 106446900 FLJ36031 | PIK3CG C 0.1323 0.3697 0.2061 0.6631 0.0008448 5 RS7446493 6974432 POLS | LOC442132 G 0.1029 1.905 1.3 2.791 0.0009471 6 RS2817457 156943434 NOX3 | ARID1B A 0.4774 1.648 1.229 2.21 0.0008462 1 RS3907135 106526262 LOC642337 | LOC126987 G 0.4926 1.628 1.219 2.173 0.0009488 24 1 RS3006993 114570444 OLFML3 | SYT6 A 0.2696 1.671 1.236 2.258 0.0008466 2 RS4588165 35326722 MYADML | LOC100288911 A 0.414 1.631 1.22 2.18 0.0009507 11 RS7128466 120491438 ARHGEF12 | GRIK4 A 0.1437 0.4015 0.2349 0.6863 0.000849 5 RS2401793 13332878 CTNND2 | DNAH5 C 0.1123 1.952 1.313 2.904 0.0009537 25 11 RS11025673 20663967 SLC6A5 G 0.3812 1.6 1.214 2.109 0.000851 3 RS11709505 53749498 CACNA1D G 0.2328 1.684 1.236 2.294 0.0009593 3 RS1523759 77604242 ROBO2 G 0.1711 0.4267 0.2586 0.7039 0.0008541 2 RS9288433 212407077 ERBB4 A 0.3101 1.634 1.221 2.187 0.000961 26 12 RS10746041 106108279 C12orf75 | NUAK1 G 0.2264 1.711 1.248 2.347 0.0008554 10 RS4751800 122636730 BRWD2 G 0.1559 0.4236 0.2543 0.7054 0.0009626 6 RS4895999 134147153 MGC34034 G 0.3805 1.614 1.218 2.139 0.0008575 23 RS6639393 4052872 LOC729162 | NLGN4X G 0.2684 1.885 1.294 2.747 0.0009636 27 3 RS10804937 103308760 LOC100287880 | ALCAM A 0.3184 1.624 1.221 2.16 0.0008578 20 RS6109068 11588817 JAG1 | LOC728573 G 0.1211 0.3592 0.1955 0.6599 0.0009683 13 RS9557607 101950392 NALCN G 0.03036 2.92 1.555 5.484 0.0008589 7 RS7788775 141817070 LOC93432 A 0.2318 0.5088 0.3406 0.7601 0.0009702 28 10 RS10826503 29010588 BAMBI | LOC729601 A 0.3866 1.636 1.225 2.185 0.0008601 15 RS1426079 70044495 LOC145837 | C15orf50 A 0.4123 1.63 1.219 2.178 0.0009716 10 RS619030 11206281 CUGBP2 A 0.2328 1.712 1.248 2.349 0.0008633 7 RS2613585 111186693 IMMP2L G 0.1164 0.317 0.1601 0.6274 0.0009726 29 18 RS151472 3447871 TGIF1 A 0.2908 1.652 1.229 2.219 0.0008653 9 RS1889321 113300835 SVEP1 G 0.2682 1.65 1.225 2.221 0.0009738 1 RS476093 181787259 CACNA1E | ZNF648 C 0.2446 1.683 1.239 2.285 0.0008663 10 RS7094645 72796318 PCBD1 | UNC5B A 0.2481 0.5231 0.3559 0.7689 0.0009773 30 1 RS12130076 240918105 GREM2 | LOC645939 G 0.1238 0.3612 0.1983 0.6577 0.0008678 3 RS7633016 46728659 ALS2CL A 0.3192 0.5552 0.3913 0.7877 0.0009774 1 RS2169089 5088617 AJAP1 | LOC100287877 A 0.2328 0.5191 0.3529 0.7635 0.0008679 18 RS6506772 75918499 GALR1 | LOC100132713 G 0.254 0.5182 0.3505 0.7662 0.000983 31 20 RS6032678 44777295 CD40 | CDH22 A 0.4514 1.631 1.223 2.176 0.000871 3 RS13084580 39188182 CSRNP1 A 0.1111 1.906 1.299 2.798 0.000987 12 RS4766019 3089354 TEAD4 A 0.123 0.3318 0.1733 0.6353 0.0008719 7 RS10276017 140979971 LOC100129514 | LOC100130169 A 0.04082 2.507 1.451 4.333 0.0009892 32 2 RS16985278 18734072 KCNS3 | RDH14 A 0.2635 1.684 1.239 2.288 0.0008739 13 RS7323413 55363417 LOC100287887 | PRR20 C 0.03745 3.145 1.59 6.221 0.0009912 3 RS1163368 108946466 C3orf66 | DPPA2 A 0.3219 1.621 1.22 2.155 0.0008751 9 RS6415721 14964952 FREM1 | LOC389705 G 0.319 0.571 0.409 0.797 0.0009913 33 7 RS11971933 111717031 DOCK4 G 0.09615 2.053 1.344 3.135 0.0008756 20 RS11086320 49687647 MULTIPLE_GENES:3755,100128598 G 0.3013 1.662 1.228 2.249 0.0009953
18 RS7506330 9177894 ANKRD12 G 0.4774 0.6034 0.4481 0.8125 0.0008765 4 RS1995917 24207765 PPARGC1A | LOC729175 G 0.06545 2.127 1.357 3.333 0.0009991 http://bmjopen.bmj.com/ 34 6 RS9442693 72386001 C6orf155 | RIMS1 A 0.05027 2.582 1.476 4.514 0.0008798 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 24 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 C. SUPPLEMENTAL REFERENCES 3 4 1. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components 5 analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38:904- 6 909 7 2. Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by 8 genome-wide association. Genomics. 2008;92:265-272 9 3. Liu KY, Muehlschlegel JD, Perry TE, Fox AA, Collard CD, Body SC, Shernan SK. Common 10 genetic variants on chromosome 9p21 predict perioperative myocardial injury after coronary 11 artery bypass graft surgery. J Thorac Cardiovasc Surg. 2010;139:483-488, 488 e481-482 12 4. Muehlschlegel JD, Liu KY, Perry TE, Fox AA, Collard CD, Shernan SK, Body SC. Chromosome 13 14 9p21 variant predicts mortality after coronary artery bypass graft surgery. Circulation. 15 2010;122:S60For-65 peer review only 16 5. Kraft P. Efficient two-stage genome-wide association designs based on false positive report 17 probabilities. Pac Symp Biocomput. 2006:523-534 18 6. Zuo Y, Zou G, Wang J, Zhao H, Liang H. Optimal two-stage design for case-control association 19 analysis incorporating genotyping errors. Ann Hum Genet. 2008;72:375-387 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41
42 on October 1, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 25 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 52 BMJ Open
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 STROBE Statement—checklist of items that should be included in reports of observational studies 3 4 Item 5 No Recommendation 6 Title and abstract 7 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 8 (b) Provide in the abstract an informative and balanced summary of what was done 9 and what was found 10 Introduction 11 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 13 Objectives 3 State specific objectives, including any prespecified hypotheses 14 Methods 15 For peer review only 16 Study design 4 Present key elements of study design early in the paper 17 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 18 exposure, follow-up, and data collection 19 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of 20 21 selection of participants. Describe methods of follow-up 22 Case-control study—Give the eligibility criteria, and the sources and methods of 23 case ascertainment and control selection. Give the rationale for the choice of cases 24 and controls 25 26 Cross-sectional study—Give the eligibility criteria, and the sources and methods of 27 selection of participants 28 (b) Cohort study—For matched studies, give matching criteria and number of 29 exposed and unexposed 30 Case-control study—For matched studies, give matching criteria and the number of 31 32 controls per case 33 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect http://bmjopen.bmj.com/ 34 modifiers. Give diagnostic criteria, if applicable 35 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 36 37 measurement assessment (measurement). Describe comparability of assessment methods if there 38 is more than one group 39 Bias 9 Describe any efforts to address potential sources of bias 40 Study size 10 Explain how the study size was arrived at 41
42 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, on October 1, 2021 by guest. Protected copyright. 43 describe which groupings were chosen and why 44 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 45 (b) Describe any methods used to examine subgroups and interactions 46 (c) Explain how missing data were addressed 47 48 (d) Cohort study—If applicable, explain how loss to follow-up was addressed 49 Case-control study—If applicable, explain how matching of cases and controls was 50 addressed 51 Cross-sectional study—If applicable, describe analytical methods taking account of 52 53 sampling strategy 54 (e) Describe any sensitivity analyses 55 Continued on next page 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 52 of 52
1 BMJ Open: first published as 10.1136/bmjopen-2014-006920 on 6 May 2015. Downloaded from 2 3 Results 4 5 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, 6 examined for eligibility, confirmed eligible, included in the study, completing follow-up, and 7 analysed 8 (b) Give reasons for non-participation at each stage 9 (c) Consider use of a flow diagram 10 11 Descriptive 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information 12 data on exposures and potential confounders 13 (b) Indicate number of participants with missing data for each variable of interest 14 (c) Cohort study—Summarise follow-up time (eg, average and total amount) 15 For peer review only 16 Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 17 Case-control study—Report numbers in each exposure category, or summary measures of 18 exposure 19 Cross-sectional study—Report numbers of outcome events or summary measures 20 21 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their 22 precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and 23 why they were included 24 (b) Report category boundaries when continuous variables were categorized 25 26 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful 27 time period 28 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 29 analyses 30 31 Discussion 32 Key results 18 Summarise key results with reference to study objectives 33 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. 34 http://bmjopen.bmj.com/ Discuss both direction and magnitude of any potential bias 35 36 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity 37 of analyses, results from similar studies, and other relevant evidence 38 Generalisability 21 Discuss the generalisability (external validity) of the study results 39 Other information 40 41 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable,
42 for the original study on which the present article is based on October 1, 2021 by guest. Protected copyright. 43 44 *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and 45 46 unexposed groups in cohort and cross-sectional studies. 47 48 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 49 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 50 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 51 52 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 53 available at www.strobe-statement.org. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2