Original Article Targeted Sequencing Reveals KLF4 Gene Associated with NSCL/P in Western Han Chinese

Int J Clin Exp Pathol 2019;12(10):3894-3900 www.ijcep.com /ISSN:1936-2625/IJCEP0098598

Original Article Targeted sequencing reveals KLF4 gene associated with NSCL/P in Western Han Chinese

Yansong Lin1, Jiayu Shi2, Bing Shi1, Zhonglin Jia1

1State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & Department of Cleft Lip and Palate, West China Hospital of Stomatology, Sichuan University, Chengdu, Sichuan, China; 2Division of Growth and Development and Section of Orthodontics, School of Dentistry, University of California, Los Angeles, USA Received June 17, 2019; Accepted August 29, 2019; Epub October 1, 2019; Published October 15, 2019

Abstract: Objective: Non-syndromic cleft lip with or without palate (NSCL/P) is one of the most common congenital birth defects affected by both gene and environmental modifiers. Recent studies revealed that KLF4 played an important role during craniofacial development. This study aimed to evaluate the association between KLF4 gene variations and NSCL/P in a Western Han Chinese population. Methods: In this study, 159 unrelated cases with NSCL/P were recruited. We used targeted region sequencing around KLF4 gene, and single variation association analysis and gene-based burden analysis were then performed. Results: 116 SNPs and 30 indels were found. Burden analysis showed no statistical significance. Association analysis results showed rs6477574 (P = 0.014 and OR = 1.58), rs112488950 (P = 0.0024 and OR = 1.98) and rs111357138 (P = 0.0025 and OR = 1.96) were associated with NSCL/P. Conclusions: The results revealed that KLF4 variations were associated with NSCL/P in a Western Han Chinese population, providing a new reference for future genetic study of NSCL/P.

Keywords: Non-syndromic cleft lip with or without palate, KLF4, association analysis

Introduction associated common SNPs. According to the hypothesis that causal genes could harbor an Non-syndromic cleft lip with or without palate excess of rare functional mutations, re-se- (NSCL/P) is one of the most common congeni- quencing studies have been performed and fur- tal birth defects worldwide. The average inci- ther provided powerful evidence for the patho- dence worldwide is 0.8 per thousand live births physiologic relevance of given genes. As an [1], while the incidence in China is higher example, re-sequencing the 1p22 loci, which than that, at 1.38 per thousand live births [2]. was reported by Beaty et al. (2010) based on NSCL/P can cause facial deformity, and also significant SNPs within ABCA4, provided strong affects speaking, feeding and hearing, eventu- support for the hypothesis that the downstre- ally results in social and psychological prob- am gene ARHGAP29 was the susceptible gene lems [3]. Although surgical correction techni- for NSCL/P (Leslie et al., 2012). Environment ques are improving, they can only cure part of risk factors like exposure to smoking, and drink- the defect. NSCL/P still poses a great burden ing during early pregnancy also have been con- to families and society. So elucidating the etiol- firmed [4]. ogy and pathogenesis of NSCL/P is of great significance for the prevention of this disorder [4]. KLF4 belongs to the Krüppel-like factor family, it is a zinc-finger-containing transcriptional fac- It has been suggested that NSCL/P is affected tor that regulates a diverse array of cellular pro- by both gene and environment factors. Genetic cesses such as proliferation, differentiation, studies have attempted to elucidate the genet- and apoptosis, and it is essential to reprogram ic background of it, and dozens of susceptible somatic cells into induced pluripotent stem genes have been reported to contribute to cells [10-14]. But KLF4 functions differently in NSCL/P, including IRF6, MSX1, TGF-α, FGFR1, different tissue microenvironments [15]. Liu et and SPRY2 [5-9]; nearly all of these genes are al. found the NSCL/P patients carried KLF4 va- characterized with the presence of strongly riations, and revealed that KLF4 is required for Klf4 gene is associated with NSCL/P

Table 1. Type of non-syndromic cleft lip with or without palate RNA/Protein contamination we- Sex Unilateral Family re verified by electrophoresis on Cleft type Bilateral Male Female Unknown Right Left history 1% agarose gels. 2) DNA purity and concentration were iden- NSCLO 46 33 0 29 37 13 0 tified by NanoPhotometer® spe- NSCLP 56 21 3 27 29 24 2 ctrophotometer (IMPLEN, CA, NSCL/P 102 54 3 56 66 37 2 USA) (OD260/OD280). OD value Total 204 108 6 112 132 74 4 of qualified sample is between Note: NSCL/P, non-syndromic cleft lip with or without cleft palate (NSCLO & 1.8-2.0. NSCLP); NSCLP, non-syndromic cleft lip with cleft palate; NSCLO, non-syndromic cleft lip only. We picked haplotype region spa- nning chr9:110237126-11026- the differentiation of murine oral periderm, and 2840 based on LD structure in CHB/JPT is directly downstream of IRF6 during oral peri- HapMap around KLF4 gene for deep sequenc- derm differentiation [16]. But there is no other ing (Figure 1). direct evidence to show if KLF4 variations were associated with NSCL/P, let alone in a Chinese NGS-library preparation and sequencing Han population. The targeted sequence was efficiently captured In this study, we performed targeted sequenc- and enriched from 1.0 μg genomic DNA using to investigate the role of KLF4 gene in the ing Agilent liquid capture system (Agilent etiology of NSCL/P. SureSelectXT Custom Kit) according to the manufacturer’s protocol. DNA library was se- Materials and methods quenced on Illumina Hiseq 4000 for paired-end 150 bp reads. Subjects Bioinformatics analysis This study recruited 159 unrelated cases with NSCL/P from Western Han Chinese population. All the downstream bioinformatics analyses are All patients were clinically examined in the based on the high quality clean data. Valid Department of Cleft Lip and Palate in West sequencing data is mapped to the reference China Hospital of Stomatology, Sichuan Uni- genome (GRCh37/hg19) by Burrows-Wheeler versity. All probands underwent clinical genetic Aligner (BWA) software to get the original map- evaluations to exclude congenital anomalies ping result in BAM format. Subsequently, other than NSCL/P. The patients were further Samtools and Picard are respectively utilized to sub-phenotyped in the basis of 2 categories of sort bam files and mark duplicate reads, to gen- severity (Table 1). Human subject study proto- erate a final bam file. Samtools mpileup and cols were reviewed and approved by the institu- bcftools are used to do variation calling and tional review board (IRB) of West China Hospital identify SNP, indels. ANNOVAR is performed to of Stomatology, Sichuan University in 2016 do annotation for VCF (Variation Call Format) (WCHSIRB-D-2016-012R1). All the participants file. The variation position, variation type, con- or their guardians read and signed the informed servative prediction and other information are consent before enrolling into this study. 542 obtained at this step through a variety of data- control data, used in the first phase association bases, such as dbSNP, 1000 Genome, ExAC, analysis, were from Novogene internal data- CADD and HGMD. For exonic variations, gene base (http://www.novogene.com/). transcript annotation databases, such as Con- sensus CDS, RefSeq, Ensembl and UCSC, are DNA collected and selection of targeted re- also applied for annotation to determine amino gions acid alternation. Finally, PolyPhen-2, SIFT, Mu- tationTaster, CADD were used to perform func- Peripheral blood samples were collected from tional prediction for the nonsynonymous SNVs. all the participants and their guardians. DNA was collected by salting out method from the Statistical analysis peripheral blood and then stored in Tris-EDTA buffer. The quality of isolated genomic DNA was In this study, we used the 542 normal individu- verified by using these two methods in combi- als WGS data from Novogene internal database nation: 1) DNA degradation and suspected (http://www.novogene.com/) as controls. Har-

3895 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P

Figure 1. Targeted sequencing region of KLF4 hg18. dy-Weinberg equilibrium (HWE) and the minor Results allele frequency (MAF) among the unaffected controls at each common variation was asse- Sequencing summary ssed; association analysis for common SNVs Of the 159 NSCL/P patients, in total we identi- was assessed by PLINK. For rare variations, we fied 116 SNVs and 30 indels. 73 SNVs and 10 included non-synonymous variations with data- indels exist in 1000 Genome database. We base MAF threshold (MAF < 0.01 in CHB & CHS selected the variations with call rate > 95% and from 1000 Genome database and Novogene split them into two groups, common variations internal database, MAF < 0.001 in GnomAD) (with minor allele frequency, MAF ≥ 0.01) and and bioinformatics prediction as harmful varia- rare variations (MAF < 0.01). tion by two of the four softwares (SIFT, Polyph- en, MutationTaster and CADD) to do the gene- Association analysis based burden analysis by two-tailed Fisher ex- act test by PLINK. Using the Bonferroni meth- To further investigate whether the common va- od the alpha for threshold would be P < 0.016. riations at KLF4 were associated with NSCL/

3896 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P

uniquely associated with NS- CLP (Figure 2 and Table 2).

Gene-based burden analysis

To know whether the KLF4 gene is associated with NSCL/ P among Western Han Chinese, we performed the gene-based burden analysis for rare variations. The results showed no statistical significance (Supplementary Table 1).

Discussions

NSCL/P is a multifactorial congenital birth defect thought to be affected by both environmental and genetic factors. The morphogenesis underlying it is partially elucidated. It is known to be caused by the fail- ure of fusion of palatal shelves during embryogenesis. Medial edge epithelia (MEE) is on the surface of palatal shelves, and it is composed of basal cell layer and the periderm layer. During palatogenesis, upon adhesion of the pair of palate shelves, the MEE cells undergo epithelial-mesenchymal trans- formation (EMT) to allow the shelf to fuse with the opposite one [17, 18]. The MEE prevents the pathological fusion between palatal shelves and man- dibular or lingual epithelia whi- Figure 2. LD plots of the association results for common variations at KLF4 ch leads to cleft palate [19]. in NSCL/P. Liu et al. [16] found that KLF acted directly downstream of P, we conducted association analysis on varia- the known risk factor of cleft palate, IRF6 [20], tions in unaffected controls with HWE (Hardy- and affected the differentiation of periderm. Weinberg equilibrium) P value > 0.05. Also, there is evidence showing that KLF4 is critical during EMT, but not in the process of The results showed that rs6477574 is signifi- palatogenesis [21]. It is necessary to further cantly associated with NSCL/P (P = 0.0097 and elucidate the mechanism underlying, and the OR = 1.43), which should be driven by NSCLP (P role of the gene regulatory networks including = 0.014 and OR = 1.58), since it has lower p members of KLF4 and IRF6. value in NSCLP than that in NSCLO. rs- 112488950 (P = 0.0024 and OR = 1.98), In this study, we performed targeted region rs111357138 (P = 0.0025 and OR = 1.96) and sequencings around the KLF4 gene, and found rs10739247 (P = 0.0089 and OR = 1.43) were two SNPs rs188071068 (NM_001314052:

3897 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P

Table 2. Association analysis results of common variants Case Control Cleft type SNP Position Function Ref Alt AF P OR Alt Ref Alt Ref NSCL/P rs6477574 110255461 intergenic A G 69.1% 217 97 662 422 0.0097 1.43 rs112488950 110239297 intergenic C T 17.6% 56 262 130 954 0.011 1.57 rs10739247 110255925 intergenic A T 68.2% 217 101 656 428 0.012 1.4 rs111357138 110238868 intergenic G T 17.6% 56 262 131 953 0.014 1.55 NSCLO rs10978885 110257762 intergenic A C 25.3% 40 118 209 875 0.088 1.42 rs10978889 110259639 intergenic A T 25.3% 40 118 209 875 0.088 1.42 NSCLP rs112488950 110239297 intergenic C T 21.3% 34 126 130 954 0.0024 1.98 rs111357138 110238868 intergenic G T 21.3% 34 126 131 953 0.0025 1.96 rs10739247 110255925 intergenic A T 71.3% 114 46 656 428 0.0089 1.62 rs6477574 110255461 intergenic A G 71.3% 114 46 662 422 0.014 1.58 rs373655937 110250376 exonic T C 1.3% 2 158 0 1084 0.016 Inf rs192594456 110255183 intergenic A G 1.3% 2 158 0 1084 0.016 Inf rs10739246 110242366 intergenic A G 71.3% 114 46 674 410 0.028 1.51 rs10759240 110242157 intergenic A C 71.3% 114 46 675 409 0.028 1.5 rs10739245 110242349 intergenic A G 71.3% 114 46 675 409 0.028 1.5 rs73523791 110238817 intergenic G C 53.8% 86 74 488 596 0.042 1.42 rs575633365 110256714 intergenic C T 1.3% 2 158 1 1083 0.045 13.66 Note: Ref, reference allele; Alt, alternate allele; AF, allele frequency; P, p value; OR, odds ratio; NSCL/P, non-syndromic cleft lip with or without cleft palate (NSCLO & NSCLP); NSCLO, non-syndromic cleft lip only; NSCLP, non-syndromic cleft lip with cleft palate. We only list the SNPs with p-values less than 0.1 in this table. c.C1195A:p.P399T) and rs373655937 (NM_ rs112488950 could alter two transcription fac- 004235: c.A299G:p.N100S) were missense, tor binding (TFB) sites, including BDP1_disc1 but they are not novel and not predicted as del- (score: -1.8 to -13.8) and NRSF_disc5 (score: eterious by SIFT/Polyphen2. 11.6 to 8.2); T allele at rs111357138 could alter three TFB sites, ERalpha-a_disc1 (score: To investigate the role of KLF4 in NSCL/P, we 11.7 to 7.2), GCNF (score: -7.8 to -7.4) and did single variation association analysis and RXRA_known5 (score: -8.3 to 4.7); and T allele gene-based burden analysis. The burden at rs1073924 could alter Mtf1_1 TFB site analysis showed no significant association (score: -9.5 to 7.5). Thus allele T at these three (Supplementary Table 1), but according to sin- SNPs in KLF4 might modulate the correspond- gle variation association analysis, KLF4 gene ing transcription and expression to participate might be associated with NSCL/P (Lowest P = in the etiology of NSCL/P. The BDP1, combined 0.0024) (Figure 2 and Table 2). We confirmed with BRF1 and TBP, constitutes TFIIIB, which is that KLF4 variations are associated with required for initiating RNA polymerase III tran- NSCL/P. The minor allele frequency (MAF) of scription [22]. The neuron restrictive silencer the significant SNP rs6477574, rs112488950, factor (NRSF) is known as transcriptional rs111357138 and rs1073924 were consistent repressor of neuronal phenotype, affecting the with the data of CHB & CHS from 1000 Genome growth and development of the nervous sys- database, confirming the representativeness tem, and the dysregulation of NRSF may lead to of the data. In conclusion, the present study neuronal cell pathologic states [23]. ER alpha-a confirmed thatKLF4 gene was a part of the eti- is a promoter of pro-oncogenic signaling path- ology of NSCL/P, and that parents carrying the ways, and ERα positivity is showed in more risk variation of KLF4 could have a greater risk than 70% of breast cancer [24, 25]. GCNF to give birth to a baby with cleft among Western stands for germ cell nuclear factor, and it is Han Chinese. Although these four variations active in testes and ovaries and is essential for are located in the non-coding region, they could reproduction in female mice [26]. As a member still affect the level and form of mRNA tran- of the nuclear receptor superfamily, RXRA is scripts. Motif analysis revealed that allele T at involved in various cellular process including

3898 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P cell differentiation, cell death and cellular se- and environmental influences. Nat Rev Genet nescence associated with calcium signaling 2011; 12: 167-178. [27, 28]. Mtf1 is a zinc finger transcription fac- [5] Mangold E, Ludwig KU and Nöthen MM. Break- tor that activates downstream gene MT1 and throughs in the genetics of orofacial clefting. Trends Mol Med 2011; 17: 725-733. MMPs to promote cell survival, and Ji et al. [29] [6] van den Boogaard MJ, Dorland M, Beemer FA had proven it to be associated with the epithe- and van Amstel HK. Msx1 mutation is associ- lial to mesenchymal transition during ovarian ated with orofacial clefting and tooth agenesis cancer metastasis. No study has shown it to be in humans. Nat Genet 2000; 24: 342-343. involved in CLP pathogenesis. [7] Yu Y, Zuo X, He M, Gao J, Fu Y, Qin C, Meng L, Wang W, Song Y, Cheng Y, Zhou F, Chen G, There are also important limitations in this Zheng X, Wang X, Liang B, Zhu Z, Fu X, Sheng study, including a very small sample size and Y, Hao J, Liu Z, Yan H, Mangold E, Ruczinski I, lack of functional studies. Further study is still Liu J, Marazita ML, Ludwig KU, Beaty TH, necessary to elucidate the role of gene regula- Zhang X, Sun L and Bian Z. Genome-wide anal- tory networks including KLF4. yses of non-syndromic cleft lip with palate identify 14 novel loci and genetic heterogene- Acknowledgements ity. Nat Commun 2017; 8: 14364. [8] Proetzel G, Pawlowski SA, Wiles MV, Yin M, Boivin GP, Howles PN, Ding J, Ferguson MW The authors thank all the participants who and Doetschman T. Transforming growth donated samples in this study. This project was factor-β3 is required for secondary palate fu- supported by the National Key R&D Program of sion. Nat Genet 1995; 11: 409-414. China (2016YFC0905200) and the National [9] Riley BM, Mansilla MA, Ma J, Daack-Hirsch S, Science Funds of China (No. 81600849). Maher BS, Raffensperger LM, Russo ET, Vieira AR, Dodé C, Mohammadi M, Marazita ML and Disclosure of conflict of interest Murray JC. Impaired fgf signaling contributes to cleft lip and palate. Proc Natl Acad Sci U S A None. 2007; 104: 4512-4517. [10] Qi XT, Li YL, Zhang YQ, Xu T, Lu B, Fang L, Gao Address correspondence to: Drs. Zhonglin Jia and JQ, Yu LS, Zhu DF, Yang B, He QJ and Ying MD. Bing Shi, State Key Laboratory of Oral Diseases & Klf4 functions as an oncogene in promoting National Clinical Research Center for Oral Diseases cancer stem cell-like characteristics in osteo- & Department of Cleft Lip and Palate, West China sarcoma cells. Acta Pharmacol Sin 2019; 40: 546-555. Hospital of Stomatology, Sichuan University, No. 14, [11] Dhaliwal NK, Miri K, Davidson S, Tamim El Jar- 3rd Section, Renmin Nan Road, Chengdu, Sichuan, kass H and Mitchell JA. KLF4 nuclear export China. Tel: +86-02885503462; Fax: +86-02885- requires ERK activation and initiates exit from 502848; E-mail: [email protected] (ZLJ); shib- naive pluripotency. Stem Cell Reports 2018; [email protected] (BS) 32: 1308-1323. [12] Ye B, Liu B, Hao L, Zhu X, Yang L, Wang S, Xia References P, Du Y, Meng S, Huang G, Qin X, Wang Y, Yan X, Li C, Hao J, Zhu P, He L, Tian Y and Fan Z. Klf4 [1] Tanaka SA, Mahabir RC, Jupiter DC and Mene- glutamylation is required for cell reprogram- zes JM. Updating the epidemiology of cleft lip ming and early embryonic development in with or without cleft palate. Plast Reconstr mice. Nat Commun 2018; 9: 1261. Surg 2012; 129: 511e-518e. [13] Fujii Y, Kakegawa M, Koide H, Akagi T and Yo- [2] Fan D, Wu S, Liu L, Xia Q, Tian G, Wang W, Ye kota T. Zfp296 is a novel klf4-interacting pro- S, Wang L, Rao J, Yang X, Yu Z, Xin L, Li S, Duan tein and functions as a negative regulator. Bio- Z, Zhang T, Wu S, Guo X and Liu Z. Prevalence chem Biophys Res Commun 2013; 441: of non-syndromic orofacial clefts: based on 411-417. 15,094,978 chinese perinatal infants. Onco- [14] Park CS, Lewis A, Chen T and Lacorazza D. target 2018; 9: 13981-13990. Concise review: regulation of self-renewal in [3] Lu X, Yu W, Tao Y, Zhao PL, Li K, Tang LJ, Zheng normal and malignant hematopoietic stem JY and Li LX. Contribution of transforming cells by Krüppel-Like factor 4. Stem Cells growth factor? Polymorphisms to nonsyndrom- Transl Med 2019; 8: 568-574. ic orofacial clefts: a huge review and meta- [15] Yadav SS, Nair RR and Yadava PK. Klf4 signal- analysis. Am J Epidemiol 2015; 179: 267-281. ling in carcinogenesis and epigenetic regula- [4] Dixon MJ, Marazita ML, Beaty TH and Murray tion of htert. Med Hypotheses 2018; 115: 50- JC. Cleft lip and palate: understanding genetic 53.

3899 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P

[16] Liu H, Leslie EJ, Jia Z, Smith T, Eshete M, Bu- [24] Nassa G, Salvati A, Tarallo R, Gigantino V, Alex- tali A, Dunnwald M, Murray J and Cornell RA. androva E, Memoli D, Sellitto A, Rizzo F, Malan- Irf6 directly regulates homologues klf17 in ze- ga D, Mirante T, Morelli E, Nees M, Åkerfelt M, brafish periderm and klf4 in murine oral epi- Kangaspeska S, Nyman TA, Milanesi L, Giurato thelium, and dominant-negative klf4 variations G and Weisz A. Inhibition of histone methyl- are present in patients with cleft lip and pal- transferase DOT1L silences ERα gene and ate. Hum Mol Genet 2016; 25: 766-776. blocks proliferation of antiestrogen-resistant [17] Cuervo R and Covarrubias L. Death is the ma- breast cancer cells. Sci Adv 2019; 5: jor fate of medial edge epithelial cells and the eaav5590. cause of basal lamina degradation during pal- [25] Lettlova S, Brynychova V, Blecha J, Vrana D, atogenesis. Development 2004; 131: 15-24. Vondrusova M, Soucek P and Truksa J. MiR- [18] Zhang YD, Dong SY and Huang HZ. Inhibition of 301a-3p suppresses estrogen signaling by di- periderm removal in all-trans retinoic acid-in- rectly inhibiting ESR1 in ERα positive breast duced cleft palate in mice. Exp Ther Med 2017; cancer. Cell Physiol Biochem 2018; 46: 2601- 14: 3393-3398. 2615. [19] Hammond NL, Dixon J and Dixon MJ. Periderm: [26] Wang H, Wang X, Xu X, Kyba M and Cooney AJ. life-cycle and function during orofacial and epi- Germ cell nuclear factor (GCNF) represses dermal development. Semin Cell Dev Biol Oct4 expression and globally modulates gene 2019; 91: 75-83. expression in human embryonic stem (hES) [20] Takahashi K, Tanabe K, Ohnuki M, Narita M, cells. J Biol Chem 2016; 291: 8644-52. Ichisaka T, Tomoda K and Yamanaka S. Induc- [27] Evans RM and Mangelsdorf DJ. Nuclear recep- tion of pluripotent stem cells from adult hu- tors, RXR, and the big bang. Cell 2014; 157: man fibroblasts by defined factors. Cell 2007; 255-266. 131: 861-72. [28] Xingjie M, Marine W, Clotilde R, Ferrand M, [21] Tiwari N, Meyer-Schaller N, Arnold P, Antonia- Kirsh O, Defossez PA, Martin N and Bernard D. dis H, Pachkov M, van Nimwegen E and Chris- The nuclear receptor RXRA controls cellular se- tofori G. Klf4 is a transcriptional regulator of nescence by regulating calcium signaling. Ag- genes critical for emt, including jnk1 (mapk8). ing Cell 2018; 17: e12831. PLoS One 2013; 8: e57329. [29] Ji L, Zhao G, Zhang P, Huo W, Dong P, Watari H, [22] Gouge J, Guthertz N, Kramm K, Dergai O, Jia L, Pfeffer LM, Yue J and Zheng J. Knockout Abascal-Palacios G, Satia K, Cousin P, Hernan- of MTF1 inhibits the epithelial to mesenchymal dez N, Grohmann D and Vannini A. Molecular transition in ovarian cancer cells. J Cancer mechanisms of Bdp1 in TFIIIB assembly and 2018; 9: 4578-4585. RNA polymerase III transcription initiation. Nat Commun 2017; 8: 130. [23] Thompson R and Chan C. NRSF and its epigenetic effectors: new treatments for neurological disease. Brain Sci 2018; 8: 226.

3900 Int J Clin Exp Pathol 2019;12(10):3894-3900 Klf4 gene is associated with NSCL/P

Supplementary Table 1. Gene-based burden analysis on rare variants among NSCL/P Case Control Cleft Type Region P Ref Alt Ref Alt NSCL/P chr9:110237125-110262840 74 86 177 254 0.13 NSCLO chr9:110237125-110262840 44 36 177 254 0.11 NSCLP chr9:110237125-110262840 30 50 177 254 0.49 Note: Ref, reference allele; Alt, alternate allele; P, p value; NSCL/P, non-syndromic cleft lip with or without cleft palate (NSCLO & NSCLP); NSCLO, non-syndromic cleft lip only; NSCLP, non-syndromic cleft lip with cleft palate.