Relevance of genes with higher information gain

BLCA related genes

CD8A achieved higher information gain considering DNA methylation. The CD8 antigen is a cell surface glycoprotein found on most cytotoxic T lymphocyte mediating efficient cell-cell interaction within the immune system. The potential for tumor control of the immune response through activated CD8+ T cells has received attention (Feldmeyer et al. 2013).

MIR663A was associated with higher information gain considering DNA methylation. It was proposed that MIR663A may play an important role in the malignant progression of chordoma (Long et al. 2013). Also MIR663 may be a putative tumor suppressor gene in pediatric acute myeloid leukemia (Tao et al. 2013).

ZSCAN18 achieved higher information gain in the domain of DNA methylation. The high sensitivity and specificity of ZSCAN18 as a biomarker for cholangiocarcinoma were identified (Andresen et al. 2012). A putative tumor suppressor function of ZSCAN18 in renal cell carcinoma was also reported (Morris et al. 2011).

CLEC3B (previously TNA) was highly informative considering RNASeqV2. The CLEC3B gene encodes tetranectin and low levels of tetranectin are associated with increased risk of second-line chemoresistance in patients with ovarian cancer (Gronlund et al. 2006). In addition, a significant association between survival duration and CLEC3B level in colorectal cancer was observed (Hogdall et al. 2002).

PYGM achieved higher information gain considering RNASeqV2. The loss of heterozygosity markers mapped to chromosome band 11q13 (PYGM locus) was subsequently demonstrated in sporadic islet-cell tumors, pituitary tumors, and parathyroid tumors (Friedman et al. 1992; Thakker et al. 1989).

ADH1B was highly informative considering RNASeqV2. This gene encodes a member of the alcohol dehydrogenase family. It was reported that the earlier diagnosis age of esophageal squamous cell carcinoma (ESCC) is associated with alcohol intake and ADH1B polymorphisms (Lee et al. 2009).

XPNPEP2 was associated with higher information gain in the domain of RNASeqV2. XPNPEP2 encodes aminopeptidase P (APP) which is a membrane protein expressed on the surface of vascular endothelial and lymphoid cells of various tissues. It was indicated that aminopeptidase P acts as a peptide receptor for a breast- homing peptide (Essler and Ruoslahti 2002).

KIAA0100 was selected considering somatic mutations. This gene is overexpressed in breast carcinoma (Song et al. 2006).

ARID1A was highly informative in the domain of somatic mutations. The observations by Wiegand and colleagues (Wiegand et al. 2010) implied the role of ARID1A as a tumor suppressor because the activity of this gene is frequently disrupted in ovarian clear-cell and endometrioid carcinomas.

MUC16 (also known as CA125) was associated with higher information gain considering somatic mutations. Besides the established role as a clinically reliable diagnostic marker for ovarian cancer (Felder et al. 2014), there have been reports that MUC16 may be associated with pathological and survival outcomes in patients with bladder cancer (Manvar et al. 2010).

ELF3 achieved higher information gain in the domain of somatic mutations. ELF3 is expressed in various carcinomas and has been shown to promote the transcription of many genes implicated in cancer. Abnormal expression of ELF3 has been implicated in the lung and breast cancer (Hou et al. 2004).

CMTM2 was selected when considering the combination of CNV and DNA methylation. It was known that CMTM2 is significantly associated with colorectal cancer (Fang et al. 2012).

BOLL was highly informative in the combination of CNV and DNA methylation. Primary lung adenocarcinomas from never smokers showed significantly higher prevalence for methylation of BOLL than smokers (current and former) (Tessema et al. 2009).

DRD4 achieved higher information gain considering the combination of CNV and DNA methylation. In a study on gastric and colorectal cancers for polymorphism, DRD4 was reported as significantly associated with gastric cancer (Ikeda et al. 2008). PCP4 was selected considering the combination of CNV and RNASeqV2. An anti-apoptosis function of PCP4 in human breast cancer cells was reported. The anti-proliferative effects of PCP4 knock-down mediated through the decreased Akt phosphorylation was demonstrated (Hamada et al. 2014).

MYRIP achieved higher information gain in the domain of CNV and RNASeqV2. It was reported that methylation of CpG sites of the potential tumor suppressor, MYRIP, is associated with hepatocellular carcinoma recurrence (Yang et al. 2011).

FAM107A was highly informative considering the combination of CNV and RNASeqV2. FAM107A is a candidate tumor suppressor gene located on chromosome 3p21.1. Down-regulation of FAM107A has been observed in various cancers such as non-small-cell lung, renal cell and prostate cancers (Nakajima and Koizumi 2014).

RAB11FIP1 was associates with higher information gain in the combination of CNV and somatic mutations. The RAB-coupling protein (RAB11FIP1) is a driving force for the 8p11-12 amplicon in human breast cancer and mouse xenograft models of mammary carcinogenesis (Subramani and Alahari 2010).

MMP23B was highly informative in the domain of DNA methylation and RNASeqV2. The cellular processes related with matrix metalloproteases (MMPs) include tissue remodeling, cell proliferation, cell migration, differentiation, apoptosis, and immune response. Furthermore, MMPs contribute to tissue degradation, tumor progression, and invasion (Galea et al. 2014).

CDO1 achieved higher information gain considering the combination of DNA methylation and RNASeqV2. Because of its relation with crucial mechanisms, CDO1 is believed as a critical tumor suppressor gene. The sensitive methylation trait of CDO1 is observed in various cancers such as breast, colorectal, esophageal, lung, bladder, and gastric cancer (Yamashita et al. 2014).

ZIC5 was associated with higher information gain in the case of the combination of DNA methylation and RNASeqV2. It is known that ZIC proteins are essential for proliferation of meningeal cell progenitors and it was observed that ZIC5 transcript level in meningiomas was higher than those in whole brain or normal dura mater (Aruga et al. 2010).

PCDHA6 was selected considering the combination of DNA methylation and somatic mutations. In breast cancer, this gene is located in a genomic region of agglomerative epigenetic aberrations (Novak et al. 2008).

CMTM2 was highly informative in the combination of DNA methylation and somatic mutations. CMTM2 was one of genes exhibiting hyper-methylation in their promoter regions associated with colorectal cancer (Fang et al. 2012).

PRDM14 achieved higher information gain considering the combination of DNA methylation and somatic mutations. The aberrant expression pattern of PRDM14 was observed in various cancers including non-small cell lung cancer and breast cancer (Nishikawa et al. 2007; Zhang et al. 2013).

FOXG1 was selected as a higher information gene considering the combination of DNA methylation and somatic mutations. One of substantial mechanisms in cancer development is loss of responsiveness to the growth inhibitory effect of TGF-. It was shown that the overexpressed FOXG1 could suppress the TGF- /Smad pathway-induced p21WAF1/CIP1 expression in ovarian cancer cells (Chan et al. 2009).

KRT24 was selected in the combination of RNASeqV2 and Somatic mutations. KRT24 plays key roles in apoptosis, adhesive migratory, and inflammatory signaling. Furthermore, it was observed that KRT24 was consistently up-regulated in the mucosa of colorectal cancer (CRC) patients compared with healthy controls (Hong et al. 2007).

ITIH5 was associated with higher information gain in the combination of RNASeqV2 and somatic mutations. ITIH5 was strongly expressed in epithelial cells of normal breast and it was lost or strongly reduced in patients of invasive breast cancer. It was indicated that ITIH5 is a candidate tumor suppressor gene (Veeck et al. 2008).

ZNF695 was highly informative considering the combination of RNASeqV2 and somatic mutations. It was suggested that ZNF695 protein regulates the expression of genes involved in a mechanism concerning DNA repair and there is a possibility that cancer cells acquire proliferation ability by loss of ZNF695 function (Takahashi et al. 2015).

KIRP related genes

SDK1 was associated with higher information gain considering CNV. The sphingosine (Sph) induced apoptotic process associated with activation of caspase 3 and release of SDK1 may promote the proapoptotic effect (Suzuki et al. 2004).

IFITM10 was highly informative considering DNA methylation and the combination of CNV and DNA methylation. The observation that the siRNA knockdown of CTSD-IFITM10 was associated with a decrease in live cells implies that this fusion plays a role in breast cancer cell proliferation (Varley et al. 2014).

KLHDC7B was associated with high information gain considering DNA methylation. In breast cancer samples, the promoter region of KLHDC7B gene was hyper-methylated (Guenin et al. 2012).

BNC1 was prominent when considering DNA methylation and the combination of CNV and DNA methylation. The p53-family member, p63 is a transcription factor that influences cellular adhesion, motility, proliferation, survival and apoptosis, and has a major role in regulating epithelial stem cells. It was shown that p63 induces the expression of the basal epithelial transcription factor, BNC1 (Boldrup et al. 2012). Boldrup and colleagues (Boldrup et al. 2012) showed that BNC1 is a direct transcriptional target of p63 and up-regulation of BNC1 is a common event in squamous cell carcinomas of the head and neck. Their work identified a new transcriptional program mediated by p63 regulation of the BNC1 transcription factor in squamous cell carcinomas.

COL9A2 was highly informative considering DNA methylation and the combination of CNV and DNA methylation. It was one of aberrantly methylated genes in gastric cancer cases considering association with Epstein-Barr virus (Matsusaka et al. 2011).

CALB1 was selected when considering RNASeqV2. CALB1 was one of commonly dysregulated genes shared by urothelial and squamous carcinoma (Hansel et al. 2013). CALB1 is expressed primarily in the kidney and is involved in the regulation of the reabsorption of calcium in the distal tubule. It was suggested that CALB1 genotypes may be a risk factor for incident renal cell carcinoma among smokers (Southard et al. 2012).

TYRP1 achieved higher information gain in the case of RNASeqV2. TYRP1 mRNA expression level was indicated as a prognostic marker for melanoma (Journe et al. 2011). Tyrosinase-related proteins influence the biology of melanocytes and melanoma (Fang et al. 2002).

MOGAT2 was highly informative in the case of RNASeqV2. MOGAT2 is hyper-methylated in breast cancer (Van der Auwera et al. 2010).

MT1G reported higher information gain in the domain of RNASeqV2. MT1G transcribes isoforms of metallothioneins, a class of low molecular weight proteins with metal-binding and antioxidant properties. In several human cancers, metallothionein expression was found to correlate with cell proliferation, tumor progression, and drug resistance (Henrique et al. 2005).

CA3 was prominent when considering the combination of CNV and DNA methylation. The assessed result of CA3 in the diagnosis of malignant effusions implied its value as tumor markers (Bramwell et al. 1985).

AQP2 was associated with high information gain in the combination of CNV and RNASeqV2. AQP2 mediates the estradiol-enhanced migration, invasion, and adhesion of endometrial carcinoma cells (Zou et al. 2011).

AKR1B10 reported higher information gain in the combination of CNV and RNASeqV2. Over-expression of AKR1B10 was found to be a superior biomarker for non-small cell lung carcinoma. It is possible that AKR1B10 involves in carcinogen metabolism (Penning 2005).

PTPRO was selected in the combination of CNV and RNASeqV2. It was reported that PTPRO is often silenced by DNA hyper-methylation in cancer cell lines and functions as a tumor suppressor (Xu et al. 2008). The PTPRO gene expresses two major transcripts. The larger transcript is expressed abundantly in the brain and kidney (Motiwala et al. 2004).

TTYH3 was selected in the combination of CNV and somatic mutations. TTYH3 was identified as one of top markers for gastric cancer (Cui et al. 2011).

ACTB was highly informative in the combination of CNV and somatic mutations. TFRC and ACTB were verified as the best combination of two genes among six housekeeping genes (TFRC, GUSB, GAPDH, ACTB, HPRT1, and RPLP0) for breast cancer quantification (Majidzadeh et al. 2011).

CDH10 reported higher information gain in the combination of RNASeqV2 and somatic mutations. It was reported that the mutation status of a five-gene-signature (CDH10, COL6A3, SMAD4, TMEM132D, VCAN) could predict survival of patients with colorectal cancer (Yu et al. 2015). Supplementary Table 1. The list of pathways associated with higher information gain genes

Target Gene name Pathway name cancer Amino acid transport across the plasma membrane, Na+/Cl- dependent SLC6A6 BLCA1 neurotransmitter transporters Downstream signaling in naïve CD8+ T cells, IL12-mediated signaling events, TCR CD8A signaling in naïve CD8+ T cells PYGM Glycogen breakdown (glycogenolysis) Chromatin remodeling by hswi/snf atp-dependent complexes, Control of gene ARID1A expression by vitamin d receptor, The information processing pathway at the IFN beta enhancer CMTM2 Coregulation of Androgen receptor activity DRD4 G alpha (i) signalling events Signaling events regulated by Ret tyrosine kinase, Direct p53 effectors, ErbB2/ErbB3 signaling events, amb2 Integrin signaling, p75(NTR)-mediated signaling, ATF-2 transcription factor network, Integrin-linked kinase signaling, Angiopoietin receptor Tie2-mediated signaling, LKB1 signaling events, PDGFR- VIT alpha signaling pathway, RAC1 signaling pathway, CDC42 signaling events, Nephrin/Neph1 signaling in the kidney podocyte, RhoA signaling pathway, Signaling events mediated by Hepatocyte Growth Factor Receptor (c-Met), TRAIL signaling pathway, Nongenotropic Androgen signalingRapid glucocorticoid signaling, Rapid glucocorticoid signaling SLC7A2 Amino acid transport across the plasma membrane FOXG1 Regulation of nuclear SMAD2/3 signaling JDP2 ATF-2 transcription factor network KIRP2 TYRP1 Direct p53 effectors DRD4 G alpha (i) signalling events PTPRO Signaling events mediated by Stem cell factor receptor (c-Kit) EPO signaling pathway, HIF-1-alpha transcription factor network, HIF-2-alpha transcription factor network, Signaling events mediated by Stem cell factor receptor (c-Kit), CDC42 signaling events, CXCR4-mediated signaling events, EPHA forward signaling, RAC1 signaling pathway, RhoA signaling pathway, CXCR3- mediated signaling events, Lissencephaly gene (LIS1) in neuronal migration and EPO development, TCR signaling in naïve CD4+ T cells, LPA receptor mediated events, PLK1 signaling events, Erythropoietin mediated neuroprotection through nf-kb, Hypoxia-inducible factor in the cardivascular system, Rac1 cell motility signaling pathway, Role of pi3k subunit p85 in regulation of actin organization and cell migration Chromatin remodeling by hswi/snf atp-dependent complexes, The information ACTB processing pathway at the ifn beta enhancer PMS2CL Direct p53 effectors SLC12A3 Cation-coupled Chloride cotransporters 1Bladder Urothelial Carcinoma 2Kidney renal papillary cell carcinoma The listed pathways are selected from the NCI-Nature Pathway Interaction Database (http://pid.nci.nih.gov/index.shtml). Supplementary Table 2. Genes with significant levels of mutation and focal copy number changes in BLCA.

Genes with statistically significant levels of Genes with statistically significant focal copy mutation number changes Gene Information gain Gene Information gain TP53 0.359 CDKN2A 0.496 MLL2 0.151 E2F3 0.234 ARID1A 0.267 CCND1 0.0728 KDM6A 0.160 RB1 0.218 PIK3CA 0.207 EGFR 0.257 EP300 0.166 PPARG 0.436 CDKN1A 0.210 PVRL4 0.311 RB1 0.218 YWHAZ 0.311 ERCC2 0.0368 MDM2 0.162 FGFR3 0.333 ERBB2 0.210 STAG2 0.116 CREBBP 0.311 ERBB3 0.117 NCOR1 0.311 FBXW7 0.389 YAP1 0.261 RXRA 0.257 CCNE1 0.610 ELF3 0.309 MYC 0.311 NFE2L2 0.118 ZNF703 0.509 TSC1 0.257 FGFR3 0.333 KLF5 0.245 PTEN 0.207 TXNIP 0.117 MYCL 0.207 FOXQ1 0.371 BCL2L1 0.117 CDKN2A 0.496 RHOB 0.705 FOXA1 0.0367 PAIP1 0.116 BTG2 0.466 HRAS 0.381 ZFP36L1 0.261 RHOA 0.257 CCND3 0.257 *Above genes were produced in (TheCancerGenomeAtlasResearchNetwork 2014). Supplementary Table 3. The list of the most differentially regulated genes in renal cell carcinoma.

Gene Information gain

ZNF160 0.0493 BRD2 0.0754 SFRS18 0.102 ANKRD12 0.225 CYP3A5 0.371 SLC35E1 0.0493 PDE4C 0.405 GPATCH8 0.257 STAG3L1 0.371 BAT2D1 0.131 SMARCA4 0.0198 HNRNPL 0.0494 PTPRO 0.759 ANKRD36B 0.131 RBM25 0.160 PAX8 0.131 SRRM2 0.260 NKTR 0.191 LOC100132247 0.257 GYG1 0.223 ITFG1 0.223 VAMP3 0.191 MTFR1 0.0754 HSP90AB1 0.0754 LAPTM4B 0.0754 PNPLA4 0.0242 CRK 0.191 GNB5 0.227 MSH6 0.131 BXDC5 0.131 C19orf2 0.0493 PICALM 0.160 TSPAN3 0.102 MRPS28 0.0754 *Above genes were produced in (Beleut et al. 2012). References

Andresen K, Boberg KM, Vedeld HM, Honne H, Hektoen M, Wadsworth CA, Clausen OP, Karlsen TH, Foss A, Mathisen O et al. (2012) Novel target genes and a valid biomarker panel identified for cholangiocarcinoma. Epigenetics 7:1249-1257 Aruga J, Nozaki Y, Hatayama M, Odaka YS, Yokota N (2010) Expression of ZIC family genes in meningiomas and other brain tumors. BMC Cancer 10:79 Beleut M, Zimmermann P, Baudis M, Bruni N, Buhlmann P, Laule O, Luu VD, Gruissem W, Schraml P, Moch H (2012) Integrative genome-wide expression profiling identifies three distinct molecular subgroups of renal cell carcinoma with different patient outcome. BMC Cancer 12:310 Boldrup L, Coates PJ, Laurell G, Nylander K (2012) p63 transcriptionally regulates BNC1, a Pol I and Pol II transcription factor that regulates ribosomal biogenesis and epithelial differentiation. Eur J Cancer 48:1401-1406 Bramwell ME, Ghosh AK, Smith WD, Wiseman G, Spriggs AI, Harris H (1985) Ca2 and Ca3. New monoclonal antibodies evaluated as tumor markers in serous effusions. Cancer 56:105-110 Chan DW, Liu VW, To RM, Chiu PM, Lee WY, Yao KM, Cheung AN, Ngan HY (2009) Overexpression of FOXG1 contributes to TGF-beta resistance through inhibition of p21WAF1/CIP1 expression in ovarian cancer. Br J Cancer 101:1433-1443 Cui JA, Chen YB, Chou WC, Sun LK, Chen L, Suo JA, Ni ZH, Zhang M, Kong XX, Hoffman LL et al. (2011) An integrated transcriptomic and computational analysis for biomarker identification in gastric cancer. Nucleic Acids Res 39:1197-1207 Essler M, Ruoslahti E (2002) Molecular specialization of breast vasculature: a breast-homing phage-displayed peptide binds to aminopeptidase P in breast vasculature. Proc Natl Acad Sci U S A 99:2252-2257 Fang D, Tsuji Y, Setaluri V (2002) Selective down-regulation of tyrosinase family gene TYRP1 by inhibition of the activity of melanocyte transcription factor, MITF. Nucleic Acids Res 30:3096-3106 Fang WJ, Zheng Y, Wu LM, Ke QH, Shen H, Yuan Y, Zheng SS (2012) Genome-wide Analysis of Aberrant DNA Methylation for Identification of Potential Biomarkers in Colorectal Cancer Patients. Asian Pac J Cancer Prev 13:1917-1921 Felder M, Kapur A, Gonzalez-Bosquet J, Horibata S, Heintz J, Albrecht R, Fass L, Kaur J, Hu K, Shojaei H et al. (2014) MUC16 (CA125): tumor biomarker to cancer therapy, a work in progress. Mol Cancer 13:129 Feldmeyer L, Gaide O, Speiser DE (2013) Clinical Implications of CD8+T-Cell Infiltration in Frequent and Rare Cancers. J Invest Dermatol 133:1929-1932 Friedman E, De Marco L, Gejman PV, Norton JA, Bale AE, Aurbach GD, Spiegel AM, Marx SJ (1992) Allelic loss from chromosome 11 in parathyroid tumors. Cancer Res 52:6804-6809 Galea CA, Nguyen HM, George Chandy K, Smith BJ, Norton RS (2014) Domain structure and function of matrix metalloprotease 23 (MMP23): role in potassium channel trafficking. Cell Mol Life Sci 71:1191- 1210 Gronlund B, Hogdall EVS, Christensen IJ, Johansen JS, Norgaard-Pedersen B, Engelholm SA, Hogdall C (2006) Pre-treatment prediction of chemoresistance in second-line chemotherapy of ovarian carcinoma: value of serological tumor marker determination (tetranectin, YKL-40, CASA, CA 125). Int J Biol Markers 21:141-148 Guenin S, Mouallif M, Deplus R, Lampe X, Krusy N, Calonne E, Delbecque K, Kridelka F, Fuks F, Ennaji MM et al. (2012) Aberrant promoter methylation and expression of UTF1 during cervical carcinogenesis. PLoS One 7:e42704 Hamada T, Souda M, Yoshimura T, Sasaguri S, Hatanaka K, Tasaki T, Yoshioka T, Ohi Y, Yamada S, Tsutsui M et al. (2014) Anti-apoptotic effects of PCP4/PEP19 in human breast cancer cell lines: a novel oncotarget. Oncotarget 5:6076-6086 Hansel DE, Zhang Z, Petillo D, Teh BT (2013) Gene profiling suggests a common evolution of bladder cancer subtypes. BMC Med Genomics 6:42 Henrique R, Jeronimo C, Hoque MO, Nomoto S, Carvalho AL, Costa VL, Oliveira J, Teixeira MR, Lopes C, Sidransky D (2005) MT1G hypermethylation is associated with higher tumor stage in prostate cancer. Cancer Epidemiology Biomarkers & Prevention 14:1274-1278 Hogdall CK, Christensen IJ, Stephens RW, Sorensen S, Norgaard-Pedersen B, Nielsen HJ (2002) Serum tetranectin is an independent prognostic marker in colorectal cancer and weakly correlated with plasma suPAR, plasma PAI-1 and serum CEA. APMIS 110:630-638 Hong Y, Ho KS, Eu KW, Cheah PY (2007) A susceptibility gene set for early onset colorectal cancer that integrates diverse signaling pathways: Implication for tumorigenesis. Clin Cancer Res 13:1107-1114 Hou JW, Wilder PJ, Bernadt CT, Boer B, Neve RM, Rizzino A (2004) Transcriptional regulation of the murine Elf3 gene in embryonal carcinoma cells and their differentiated counterparts: requirement for a novel upstream regulatory region. Gene 340:123-131 Ikeda S, Sasazuki S, Natsukawa S, Shaura K, Koizumi Y, Kasuga Y, Ohnami S, Sakamoto H, Yoshida T, Iwasaki M et al. (2008) Screening of 214 single nucleotide polymorphisms in 44 candidate cancer susceptibility genes: A case-control study on gastric and colorectal cancers in the Japanese population. Am J Gastroenterol 103:1476-1487 Journe F, Id Boufker H, Van Kempen L, Galibert MD, Wiedig M, Sales F, Theunis A, Nonclercq D, Frau A, Laurent G et al. (2011) TYRP1 mRNA expression in melanoma metastases correlates with clinical outcome. Br J Cancer 105:1726-1732 Lee CH, Wu DC, Wu IC, Goan YG, Lee JM, Chou SH, Chan TF, Huang HL, Hung YH, Huang MC et al. (2009) Genetic modulation of ADH1B and ALDH2 polymorphisms with regard to alcohol and tobacco consumption for younger aged esophageal squamous cell carcinoma diagnosis. Int J Cancer 125:1134- 1142 Long C, Jiang L, Wei F, Ma C, Zhou H, Yang SM, Liu XG, Liu ZJ (2013) Integrated miRNA-mRNA Analysis Revealing the Potential Roles of miRNAs in Chordomas. PLoS One 8: e66676 Majidzadeh AK, Esmaeili R, Abdoli N (2011) TFRC and ACTB as the best reference genes to quantify Urokinase Plasminogen Activator in breast cancer. BMC Res Notes 4:215 Manvar AM, Wallen EM, Pruthi RS, Nielsen ME (2010) Prognostic value of CA 125 in transitional cell carcinoma of the bladder. Expert Rev Anticancer Ther 10:1877-1881 Matsusaka K, Kaneda A, Nagae G, Ushiku T, Kikuchi Y, Hino R, Uozaki H, Seto Y, Takada K, Aburatani H et al. (2011) Classification of Epstein-Barr virus-positive gastric cancers by definition of DNA methylation epigenotypes. Cancer Res 71:7187-7197 Morris MR, Ricketts CJ, Gentle D, McRonald F, Carli N, Khalili H, Brown M, Kishida T, Yao M, Banks RE et al. (2011) Genome-wide methylation analysis identifies epigenetically inactivated candidate tumour suppressor genes in renal cell carcinoma. Oncogene 30:1390-1401 Motiwala T, Kutay H, Ghoshal K, Bai S, Seimiya H, Tsuruo T, Suster S, Morrison C, Jacob ST (2004) Protein tyrosine phosphatase receptor-type O (PTPRO) exhibits characteristics of a candidate tumor suppressor in human lung cancer. Proc Natl Acad Sci U S A 101:13844-13849 Nakajima H, Koizumi K (2014) Family with sequence similarity 107: A family of stress responsive small proteins with diverse functions in cancer and the nervous system (Review). Biomed Rep 2:321-325 Nishikawa N, Toyota M, Suzuki H, Honma T, Fujikane T, Ohmura T, Nishidate T, Ohe-Toyota M, Maruyama R, Sonoda T et al. (2007) Gene amplification and overexpression of PRDM14 in breast cancers. Cancer Res 67:9649-9657 Novak P, Jensen T, Oshiro MM, Watts GS, Kim CJ, Futscher BW (2008) Agglomerative epigenetic aberrations are a common event in human breast cancer. Cancer Res 68:8616-8625 Penning TM (2005) AKR1B10: a new diagnostic marker of non-small cell lung carcinoma in smokers. Clin Cancer Res 11:1687-1690 Song J, Yang W, Shih Ie M, Zhang Z, Bai J (2006) Identification of BCOX1, a novel gene overexpressed in breast cancer. Biochim Biophys Acta 1760:62-69 Southard EB, Roff A, Fortugno T, Richie JP, Jr., Kaag M, Chinchilli VM, Virtamo J, Albanes D, Weinstein S, Wilson RT (2012) Lead, calcium uptake, and related genetic variants in association with renal cell carcinoma risk in a cohort of male Finnish smokers. Cancer Epidemiol Biomarkers Prev 21:191-201 Subramani D, Alahari SK (2010) Integrin-mediated function of Rab GTPases in cancer progression. Mol Cancer 9:312 Suzuki E, Handa K, Toledo MS, Hakomori S (2004) Sphingosine-dependent apoptosis: a unified concept based on multiple mechanisms operating in concert. Proc Natl Acad Sci U S A 101:14788-14793 Takahashi T, Yamahsita S, Matsuda Y, Kishino T, Nakajima T, Kushima R, Kato K, Igaki H, Tachimori Y, Osugi H et al. (2015) ZNF695 methylation predicts a response of esophageal squamous cell carcinoma to definitive chemoradiotherapy. J Cancer Res Clin Oncol 141: 453-463 Tao YF, Ni J, Lu J, Wang N, Xiao PF, Zhao WL, Wu D, Pang L, Wang J, Feng X et al. (2013) The promoter of miR-663 is hypermethylated in Chinese pediatric acute myeloid leukemia (AML). BMC Med Genet 14: 74 Tessema M, Yu YY, Stidley CA, Machida EO, Schuebel KE, Baylin SB, Belinsky SA (2009) Concomitant promoter methylation of multiple genes in lung adenocarcinomas from current, former and never smokers. Carcinogenesis 30:1132-1138 Thakker RV, Bouloux P, Wooding C, Chotai K, Broad PM, Spurr NK, Besser GM, O'Riordan JL (1989) Association of parathyroid tumors in multiple endocrine neoplasia type 1 with loss of alleles on chromosome 11. N Engl J Med 321:218-224 TheCancerGenomeAtlasResearchNetwork (2014) Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 507:315-322 Van der Auwera I, Yu W, Suo L, Van Neste L, van Dam P, Van Marck EA, Pauwels P, Vermeulen PB, Dirix LY, Van Laere SJ (2010) Array-based DNA methylation profiling for breast cancer subtype discrimination. PLoS One 5:e12616 Varley KE, Gertz J, Roberts BS, Davis NS, Bowling KM, Kirby MK, Nesmith AS, Oliver PG, Grizzle WE, Forero A et al. (2014) Recurrent read-through fusion transcripts in breast cancer. Breast Cancer Res Treat 146:287-297 Veeck J, Chorovicer M, Naami A, Breuer E, Zafrakas M, Bektas N, Durst M, Kristiansen G, Wild PJ, Hartmann A et al. (2008) The extracellular matrix protein ITIH5 is a novel prognostic marker in invasive node- negative breast cancer and its aberrant expression is caused by promoter hypermethylation. Oncogene 27:865-876 Wiegand KC, Shah SP, Al-Agha OM, Zhao Y, Tse K, Zeng T, Senz J, McConechy MK, Anglesio MS, Kalloger SE et al. (2010) ARID1A mutations in endometriosis-associated ovarian carcinomas. N Engl J Med 363:1532-1543 Xu X, Hong Y, Kong CF, Xu L, Tan J, Liang Q, Huang BQ, Lu J (2008) Protein tyrosine phosphatase receptor- type O (PTPRO) is co-regulated by E2F1 and miR-17-92. FEBS Lett 582:2850-2856 Yamashita K, Waraya M, Kim MS, Sidransky D, Katada N, Sato T, Nakamura T, Watanabe M (2014) Detection of Methylated CDO1 in Plasma of Colorectal Cancer; A PCR Study. PLoS One 9:e113546 Yang JD, Seol SY, Leem SH, Kim YH, Sun Z, Lee JS, Thorgeirsson SS, Chu IS, Roberts LR, Kang KJ (2011) Genes associated with recurrence of hepatocellular carcinoma: integrated analysis by gene expression and methylation profiling. J Korean Med Sci 26:1428-1438 Yu J, Wu WK, Li X, He J, Li XX, Ng SS, Yu C, Gao Z, Yang J, Li M et al. (2015) Novel recurrently mutated genes and a prognostic mutation signature in colorectal cancer. Gut 64: 636-645 Zhang TH, Meng L, Dong W, Shen HC, Zhang SM, Liu Q, Du JJ (2013) High expression of PRDM14 correlates with cell differentiation and is a novel prognostic marker in resected non-small cell lung cancer. Med Oncol 30: 605 Zou LB, Zhang RJ, Tan YJ, Ding GL, Shi SA, Zhang D, He RH, Liu AX, Wang TT, Leung PCK et al. (2011) Identification of Estrogen Response Element in the Aquaporin-2 Gene That Mediates Estrogen-Induced Cell Migration and Invasion in Human Endometrial Carcinoma. J Clin Endocrinol Metab 96:E1399- E1408