BR Biomedical Reports 2049-9434 2049-9442 D.A. Spandidos 10.3892/br.2018.1109 BR-0-0-1109 Articles Six novel susceptibility loci for coronary artery disease and cerebral infarction identified by longitudinal exome-wide association studies in a Japanese population YasukochiYoshiki 1 2 SakumaJun 2 3 4 TakeuchiIchiro 2 4 5 KatoKimihiko 1 6 OguriMitsutoshi 1 7 FujimakiTetsuo 8 HoribeHideki 9 YamadaYoshiji 1 2 Department of Human Functional Genomics, Advanced Science Research Promotion Center, Mie University, Tsu, Mie 514-8507, Japan Core Research for Evolutional Science and Technology, Japan Science and Technology Agency, Kawaguchi, Saitama 332-0012, Japan Computer Science Department, College of Information Science, University of Tsukuba, Tsukuba, Ibaraki 305-8573, Japan RIKEN Center for Advanced Intelligence Project, Tokyo 103-0027, Japan Department of Computer Science, Nagoya Institute of Technology, Nagoya, Aichi 466-8555, Japan Department of Internal Medicine, Meitoh Hospital, Nagoya, Aichi 465-0025, Japan Department of Cardiology, Kasugai Municipal Hospital, Kasugai, Aichi 486-8510, Japan Department of Cardiovascular Medicine, Inabe General Hospital, Inabe, Mie 511-0428, Japan Department of Cardiovascular Medicine, Gifu Prefectural Tajimi Hospital, Tajimi, Gifu 507-8522, Japan Correspondence to: Dr Yoshiki Yasukochi, Department of Human Functional Genomics, Advanced Science Research Promotion Center, Mie University, 1577 Kurima-machiya, Tsu, Mie 514-8507, Japan, E-mail: hyasukou@proof.ocn.ne.jp 08 2018 05 06 2018 9 2 123 134 26012018 31052018 Copyright: © Yasukochi et al. 2018 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

Coronary artery disease (CAD) and cerebral infarction (CI) remain major causes of morbidity and mortality in humans. Recent genome-wide association studies have identified various genetic variants associated with these diseases. However, these studies were commonly conducted in a cross-sectional manner. Therefore, the present research performed longitudinal exome-wide association studies for CAD and CI using data on ~244,000 genotyped variants and the clinical data of 6,026 Japanese individuals who had attended annual health checkups for several years (mean followed-up period, 5±3 years). Following quality controls, the significance [false discovery rate (FDR) of <0.05] of association of the diseases with 24,651 single nucleotide polymorphisms (SNPs) in 5,989 individuals for three inheritance models was tested using the generalized estimating equation model. SNPs that reached statistical significance were further screened against a threshold of approxdf (a scale of small effective sample size) of >30. The longitudinal exome-wide association studies revealed that three SNPs [rs4606855 of ADGRE3 (P=2.5×10−6; FDR=0.031; approxdf=71), rs3746414 of ZFP64 (P=5.9×10−6; FDR=0.048; approxdf=93) and rs7132908 of FAIM2 (P<2.0×10−16; FDR<4.9×10−12; approxdf=65)] were significantly associated with the prevalence of CAD. A different set of three SNPs [rs6580741 of FAM186A (P<2.0×10−16; FDR<4.9×10−12; approxdf=48), rs1324015 of LINC00400 (P<2.0×10−16; FDR<4.9×10−12; approxdf=49) and rs884205 of TNFRSF11A (P<2.0×10−16; FDR<4.9×10−12; approxdf=32)] was significantly associated with CI. The comparison of disease incidence with these SNPs demonstrated that all the minor alleles were associated with decreased susceptibility to CAD or CI. In conclusion, six novel SNPs were identified as susceptibility loci for CAD (rs4606855 of ADGRE3, rs3746414 of ZFP64, and rs7132908 of FAIM2) or CI (rs6580741 of FAM186A, rs1324015 of LINC00400, and rs884205 of TNFRSF11A).

 cerebral infarction coronary artery disease exome-wide association study generalized estimating equation longitudinal data single nucleotide polymorphism Introduction

Coronary artery disease (CAD) remains the worldwide leading cause of mortality among men and women (1). The American Heart Association reported that in 2014 there were 364,593 mortalities from CAD in the United States (2), based on 2014 mortality data from the National Center for Health Statistics (3). In Japan, the Ministry of Health, Labour and Welfare has reported that the rate of mortality per 100,000 population from acute ischemic heart diseases was 71,673 in 2015 (4). Cerebral infarction (CI) is also a serious clinical problem worldwide. In the United States, approximately 610,000 and 185,000 people experience new and recurrent stroke events each year, respectively, and it was estimated that the prevalence of silent CI ranged from 6–28% from 1993 to 2005 (2). In 2015, the number of mortalities from CI and intracerebral hemorrhage in Japanese patients was 64,523 and 32,113, respectively (4). Therefore, examination of CAD and CI susceptibility variants in Japanese individuals may be key to successful personalized prevention of these diseases.

Previous genome-wide association studies (GWASs) have identified a variety of genes and loci that confer susceptibility to CAD and CI across various ethnic groups, and have reported inter-ethnic differences of genetic contribution to these diseases (5–11). A meta-analysis using datasets of the CARDIoGRAMplusC4D, MIGen and CARDIoGRAM Exome consortia and ESP EOMI studies assessed the association of genetic variants in several chemokine receptor genes with CAD and myocardial infarction (6). The study did not identify any evidence of an association between the genetic variants and disease in large European ancestry cohorts, whereas six low frequency variants were associated with myocardial infarction in a South Asian cohort. Another meta-analysis of two independent Chinese GWASs for CAD identified four CAD-associated loci that were not present in populations of European ancestry (7). Inter-ethnic difference of disease susceptibility loci has also been reported in GWASs for ischemic stroke (i.e., CI). A GWAS for ischemic stroke in 6,341 Japanese individuals who participated in three independent population-based studies identified that cadherin EGF LAG seven-pass G-type receptor 1 (CELSR1) was a susceptibility gene for ischemic stroke (10). The association of CELSR1 with stroke has been confirmed in a Portuguese cohort (12); however, the Siblings With Ischemic Stroke Study did not demonstrate an association of 312 probands with ischemic stroke from 70 centers in the USA and Canada (13,14). The discrepancy of disease-susceptible variants among populations may be due to differences of genetic background among ethnic groups.

Although recent GWASs have identified a large number of genetic variants that confer susceptibility to cardiovascular diseases (5,7,9,14,15), they have been conducted in a cross-sectional manner that commonly measures traits at a single point in time. To address this issue, the present study examined disease progression and physiological changes in 5,989 Japanese individuals who had annual health check-ups for several years, and performed longitudinal exome-wide association studies (EWASs) to investigate novel susceptibility loci for CAD and CI.

 Materials and methods <sec> <title>Study subjects

The physiological data of 6,026 community-dwelling individuals recruited to a population-based cohort study (Inabe Health and Longevity Study) in Inabe City, Japan was traced (16–20). These individuals visited Inabe General Hospital for an annual health check-up, with a mean annual follow-up period of 5±3 years (covering April 2003 to March 2014). All participants had undergone 1 to 11 medical examinations. This cohort was denoted as the ‘Inabe cohort’ in the longitudinal EWASs. Detailed methods for the recruitment of subjects and for the collection and storage of medical examination data and genomic DNA samples have been described previously (19). Diagnostic criteria for CAD and CI have also been described previously (18).

The study protocol complied with the Declaration of Helsinki and was approved by the Committees on the Ethics of Human Research of Mie University Graduate School of Medicine and Inabe General Hospital (Inabe, Japan). Written informed consent was obtained from all subjects prior to enrollment in the Inabe Health and Longevity Study.

 Longitudinal EWAS

Infinium HumanExome-12 ver. 1.2 BeadChip and Infinium Exome-24 ver. 1.0 BeadChip (Illumina, Inc., San Diego, CA, USA) were used to genotype ~244,000 genetic variants in the Inabe cohort for longitudinal EWASs. These arrays include putative functional exonic variants selected from >12,000 individual exome and whole-genome sequences across diverse ethnic populations, including European, African, Chinese and Hispanic individuals (21). Missing genotype or phenotype data of each individual was eliminated from the analyses. Quality control of genotyping data was performed, and monomorphic sites and the following genetic variants were discarded: i) Variants contained in only one of the exome arrays used (~3.6% of all variants); ii) variants with a call rate of <97.0%; iii) variants with a minor allele frequency (MAF) of <0.05; iv) variants whose genotype distribution significantly deviated from Hardy-Weinberg equilibrium (P<0.001) in controls; and v) variants located on mitochondrial DNA or sex chromosomes.

In addition, sex specification was checked for each sample, with samples for which the sex designation in the clinical records was inconsistent with genetic sex discarded. Cryptic relatedness and duplicate samples were checked by calculation of identity by descent (IBD); all pairs of DNA samples exhibiting an IBD of >0.1875 were inspected and one sample from each pair was excluded. Population stratification in the Inabe cohort was examined by principal component analysis via the EIGENSTRAT method (22), using JMP Genomics version 6.0 (SAS Institute, Inc., Cary, NC, USA), and four outliers were removed from the longitudinal EWASs. Consequently, a total of 24,651 single nucleotide polymorphisms (SNPs) among 5,989 Japanese individuals passed quality control for the longitudinal EWASs of CAD and CI. The rearrangement of Inabe longitudinal data was conducted using R software version 3.32 (23) via RStudio version 1.0.136 (http://www.rstudio.com/) (24) and Perl script (version 5.26.2; https://www.perl.org/get.html). Using JMP Genomics, genotyping data were converted into numeric data for dominant, additive and recessive inheritance models. The dominant and recessive models were defined as ‘0, AA; 1, AB + BB’ and ‘0, AA + AB; 1, BB’ (A, major allele; B, minor allele), respectively, whereas the additive model was defined as ‘0, AA; 1, AB; 2, BB’.

 Statistical analysis

Longitudinal EWASs were conducted for 5,713 control individuals, 170 subjects with CAD, and 117 with CI (Table I). The association of the prevalence of CAD and CI with SNPs was tested by the generalized estimating equation (GEE) model (25,26) with adjustments for age, sex, body mass index (BMI), smoking and prevalence of hypertension, type 2 diabetes mellitus and dyslipidemia, using the R package ‘geepack’ (27). Since the prevalence of CAD and CI is repeated categorical data (case or control), a binomial distribution was applied for assessing the association between the categorical outcomes and SNPs in the GEE method. The waves argument was used to specify the ordering of repeated measurements within individuals. Effects of SNPs in exome arrays on CAD or CI are not independent as many SNPs are in linkage disequilibrium (LD) (13,28,29). Therefore, the false discovery rate (FDR) was calculated using the Benjamin and Hochberg method (30) to compensate for multiple comparison of genotypes with the phenotypes. An FDR of <0.05 was considered to indicate a statistically significant association.

A small effective sample size may increase the probability of generating false positives (type I errors) (31). Sitlani et al (31) recommend the use of approxdf, which is a scale of small effective sample size: Approxdf = 2 × MAF × Nindep, where Nindep is the sum of the estimated number of independent observations per person. They demonstrated that an approxdf of ≥10 could reduce type I errors. Thus, approxdf was computed by the R package ‘bosswithdf’ (31,32). To avoid the issue of false positive association in small sample sizes, a strict approxdf threshold was applied, and SNPs with approxdf ≤30 discarded.

 LD estimates and prediction of functional association for candidate loci

To survey LDs between the candidate SNPs detected in the present study and previously identified CAD- or CI-associated SNPs in JPT (Japanese in Tokyo, Japan) from the 1000 Genomes Project [http://www.1000genomes.org/ (33)], analyses were conducted using the LDproxy of LDlink [https://analysistools.nci.nih.gov/LDlink/ (34)], which is a web-based application. The association of SNPs with CAD or CI reported by previous studies was investigated using the Genome-Wide Repository of Associations Between SNPs and Phenotypes [GRASP; https://grasp.nhlbi.nih.gov/Overview.aspx (35)], DisGeNET [http://www.disgenet.org/web/DisGeNET/ (36)] and GWAS Catalogue [https://www.ebi.ac.uk/gwas/ (37)] databases. Gene-gene functional interactions were predicted using the GeneMANIA Cytoscape plugin (38–40) via Cytoscape version 3.4.0 software [http://www.cytoscape.org/ (41)].

 Results <sec> <title>Characteristics of subjects

The characteristics of subjects in the Inabe cohort with respect to longitudinal data are presented in Table I. The prevalence of hypertension, type 2 diabetes mellitus, dyslipidemia, chronic kidney disease and hyperuricemia was higher in patients with CAD or CI than in controls. The prevalence of CAD and CI was lower in females (39 CAD and 37 CI patients) than in males (131 CAD and 80 CI patients). The majority of physiological or clinical parameters examined (age, BMI, systolic and diastolic blood pressure, fasting plasma glucose level, blood hemoglobin A1c content and serum concentrations of triglycerides, creatinine, and uric acid) were higher, whereas serum concentrations of high-density lipoprotein-cholesterol and estimated glomerular filtration rate were lower, in the patients than in the controls. The serum concentration of low-density lipoprotein-cholesterol was lower in patients with CAD (2.76±0.032 mmol/l) than in controls (3.19±0.005 mmol/l). This discrepancy may be attributable to effects of lipid-lowering treatment for patients with CAD.

 Association of SNPs with longitudinal data on the prevalence of CAD

Following the quality control checks, the association between 24,651 SNPs and the prevalence of CAD in the Inabe cohort was tested by using the GEE model with adjustments for age, sex, BMI, smoking and prevalence of hypertension, type 2 diabetes mellitus and dyslipidemia. Candidate SNPs that reached statistical significance (FDR of <0.05) were additionally screened against the threshold of approxdf (>30). Analyses revealed that two SNPs [rs4606855 of ADGRE3 (P=2.5×10−6; FDR=0.031, approxdf=71) and rs3746414 of ZFP64 (P=5.9×10−6; FDR=0.048; approxdf=93)] and one SNP [rs7132908 of FAIM2 (P<2.0×10−16; FDR<4.9×10−12; approxdf=65)] were significantly associated with the prevalence of CAD in additive and recessive models, respectively (Table II).

According to the GRASP, DisGeNET and GWAS Catalogue databases, the association of the three SNPs with CAD has not been reported to date. Therefore, the three SNPs were identified as novel genetic variants that may confer susceptibility to CAD. The nucleotide substitution at rs7132908 in FAIM2 is predicted to be a silent substitution in the 3′-untranslated region, while the remaining two SNPs are predicted to alter amino acid residues (i.e., nonsynonymous substitutions), according to the NCBI dbSNP database [https://www.ncbi.nlm.nih.gov/projects/SNP/ (42)].

Genotype distributions for rs4606855, rs3746414 and rs7132908 in subjects with CAD and controls in the longitudinal EWAS are listed in Table III. In the current study, the prevalence of CAD was lower in subjects with minor alleles than in those with major alleles for all SNPs, suggesting the CC genotype of rs4606855 and the AA genotypes of rs3746414 and rs7132908 to be protective against CAD.

 Association of SNPs with longitudinal data on the prevalence of CI

In the Inabe cohort, three SNPs [rs6580741 of FAM186A (P<2.0×10−16; FDR<4.9×10−12; approxdf=48), rs1324015 of LINC00400 (P<2.0×10−16; FDR<4.9×10−12; approxdf=49), and rs884205 of TNFRSF11A (P<2.0×10−16; FDR<4.9×10−12; approxdf=32)] were significantly associated with the prevalence of CI in the recessive model, although the GEE tests in the dominant and additive models detected no significant association between SNPs and the prevalence of CI (Table II). The CI-associated SNPs detected in the recessive model were not shared with the CAD-associated SNPs. The nucleotide substitution at rs6580741 in FAM186A is predicted to alter an amino acid residue (Table II), whereas the substitutions at the other SNPs are predicted as silent, according to the NCBI dbSNP.

The rs6580741 of FAM186A is located relatively close (~464 kb) to rs7132908 of FAIM2 at chromosomal region 12q13.12, although these SNPs were related to the different diseases. The LD among these SNPs was thus examined among participants in the Inabe cohort using the R package ‘genetics’ [https://CRAN.R-project.org/package=genetics (43)]. The estimation demonstrated that these SNPs were not in LD (D'=0.045, r2=0.002), suggesting that rs7132908 and rs6580741 were independently associated with the prevalence of CAD and CI, respectively.

Genotype distributions for rs6580741, rs1324015 and rs884205 in subjects with CI and controls in the longitudinal EWAS are listed in Table IV. As with the case of CAD-associated SNPs, for all three SNPs, the prevalence of CI was lower in subjects with minor alleles than in those with major alleles. The CC genotype of rs6580741, the AA genotype of rs1324015, and the TT genotype of rs884205 may thus be protective against CI.

 Assessment of LD between the six identified SNPs and other SNPs associated with CAD or CI

The LDs between the six candidate SNPs identified in the present study and adjacent SNPs across a ~1 Mb genomic region were assessed using the LDproxy in LDlink. The LDproxy analysis indicated that 433 SNPs were in significant LD (r2≥0.5) with one of the candidate SNPs in JPT from the 1000 Genomes Project (Table V). Of the 433 SNPs, there were no SNPs previously identified to be associated with CAD or CI (P>0.01) according to the GRASP database. In addition, the LDpair analysis in LDlink using allele frequency data in East Asian populations from the 1000 Genomes Project indicated that rs1324015 associated with the prevalence of CI in the present study was not in LD with rs9533425 (D'=0.4709, r2=0.0007), previously reported to be associated with lag time (lag phase of the turbidimetric clotting assay) to fibrin clot formation (28). These results suggest that effects of the candidate SNPs on the prevalence of CAD or CI are independent.

Gene interaction network analysis. To investigate the interactive functional association, a GeneMANIA network analysis was conducted of the top ten genes (high gene-disease association scores) that have been demonstrated to be associated with CAD or CI selected from the DisGeNET database, and of the five genes (ZFP64, FAIM2, ADGRE3, FAM186A, and TNFRSF11A) identified in the present study (Fig. 1). Given that LINC00400 non-coding RNA has not been well characterized, it could not be examined. The network analysis showed that the CAD- and CI-associated genes identified in the present study have potential direct or indirect interactions with several genes previously demonstrated to be associated with CAD and CI, respectively. The network suggested that ZFP64 and FAIM2 interact with LOX and both CRP and APOB, respectively. In addition, ADGRE3 and TNFRSF11A were indicated to be co-expressed with ALOX5AP and CD40LG. The network also suggested that TNFRSF11A interacts and is co-expressed with PFKCH and ADH1B, respectively. Additionally, in the network, FAM186A was indirectly connected with CI-associated genes (NPY and F5) through TMPRSS15.

 Discussion

In the present study, genetic variants that confer susceptibility to CAD and CI were investigated. The longitudinal EWASs in a Japanese cohort identified that rs4606855 of ADGRE3, rs3746414 of ZFP64 and rs7132908 of FAIM2 were significantly associated with the prevalence of CAD, whereas rs6580741 of FAM186A, rs1324015 of LINC00400 and rs884205 of TNFRSF11A were significantly associated with the prevalence of CI. According to the databases, these SNPs may be novel susceptibility loci for CAD or CI.

The adhesion G protein-coupled receptor E3 (ADGRE3) gene is located at chromosomal region 19p13.12, and rs4606855 [G/C (E75Q)] in this gene was significantly associated with the prevalence of CAD. The mRNA of ADGRE3 is predominantly expressed in neutrophils, monocytes and macrophages, and this protein may be important for interactions between myeloid cells including neutrophils and monocytes during immune and inflammatory responses (44). Given that vascular inflammatory responses serve a role in the development of coronary atherosclerosis and thrombosis (45–47), this CAD-associated SNP may be attributable to functional alteration of ADGRE3 protein in vascular inflammation.

The ZFP64 zinc finger protein (ZFP64) gene located at 20q13.2 is expressed in various tissues and organs, according to The Human Protein Atlas [http://www.proteinatlas.org/ (48,49)]. The present longitudinal EWASs demonstrated that rs3746414 [G/A (S451N)] was significantly associated with the prevalence of CAD. ZFP64 protein may be a positive regulator of Toll-like receptor-induced innate immune responses through tumor necrosis factor (TNF)-α, interleukin-6 and interferon-β production in macrophages (50). In fact, Zfp64 knockdown mice exhibited significant reductions in the production of these cytokines (50). Thus, the association of ZFP64 with CAD may be attributable to the effect of this gene on the development of coronary atherosclerosis or thrombosis.

The Fas apoptotic inhibitory molecule 2 (FAIM2) gene is located at 12q13.12, and rs7132908 (G/A) in this gene was associated with the prevalence of CAD. FAIM2 is predominantly expressed in the brain while moderate expressions are also observed in other various tissues and organs, according to The Human Protein Atlas. AIM2 inhibits Fas-mediated apoptosis (51). The Fas cell surface death receptor gene (FAS) serves important roles in the regulation of the immune system through inducing apoptosis of autoreactive or antigen-presenting T lymphocytes (52–54). Given that FAIM2 regulates Fas-mediated apoptosis during immune responses, rs7132908 in this gene may be associated with the incidence of CAD through altering the regulation of immune or inflammatory responses.

The TNF receptor superfamily member 11a (TNFRSF11A) gene is located at chromosomal region 18q21.33, and is expressed in various tissues and organs including the brain, according to The Human Protein Atlas. In the present study, rs884205 (G/T) of TNFRSF11A was significantly associated with the prevalence of CI. TNFRSF11A and its ligand (TNFRSF11) may be important regulators of the interaction between T cells and dendritic cells that are involved in immune surveillance (55,56). In mice, TNFRSF11/TNFRSF11A (also known as RANKL/RANK) signaling could trigger inflammatory fever responses in the central nervous system (55), and may serve anti-inflammatory roles in ischemic brains (56). The rs884205 in TNFRSF11A may thus influence the development of CI.

The exact functions of proteins encoded by the family with sequence similarity 186 member A (FAM186A) and long intergenic non-protein coding RNA 400 (LINC00400) genes are unknown. However, it has been reported that rs9533425 located near LINC00400 at chromosomal region 13q14.11 demonstrated a significant association (P=1.9×10−9) with lag time to fibrin clot formation in 2,100 subjects from the Twins United Kingdom (TwinsUK) registry in stage 1 study, although the association of this SNP was not replicated in stage 2 study (28). In the present longitudinal EWASs, rs1324015 of LINC00400 was significantly associated with the prevalence of CI. According to LDpair in LDlink, rs1324015 was not in LD with rs9533425 in East Asian populations. Although the functional relevance of the candidate SNP to the development of cerebral atherosclerosis or thrombosis remains unclear, LINC00400 may be a susceptibility locus for the incidence of CI.

There were certain limitations in the present study. First, the longitudinal EWASs were conducted in only a local Japanese population, and the observed number of patients who were affected by the target diseases was not sufficient. Thus, replication of longitudinal EWASs in other Japanese populations or other ethnic groups is required to verify the association of the identified SNPs with the diseases of interest. However, to the best of our knowledge, longitudinal data for CAD and CI in other populations are unavailable at present. Second, no experiments for functional analyses were conducted in the present study. Thus, the functional relevance of the candidate SNPs identified by longitudinal EWASs to the pathogenesis of the diseases of interest remains unclear. Due to the lack of experiments for functional analyses, the association of the SNPs identified in the present study with CAD or CI should be interpreted with caution. Further functional analyses are required to clarify the present results.

In conclusion, rs4606855 of ADGRE3, rs3746414 of ZFP64 and rs7132908 of FAIM2 may be susceptibility loci for CAD. Additionally, the SNPs rs6580741 of FAM186A, rs1324015 of LINC00400 and rs884205 of TNFRSF11A may be genetic determinants of CI. All minor alleles of the six candidate SNPs exhibited an association with lower prevalence of the associated diseases, compared with the corresponding major alleles. This suggests that the minor alleles of each candidate SNP may be protective against CAD or CI.

 Acknowledgements

Not applicable.

 Funding

The present study was supported by a research grant from the Okasan Kato Culture Promotion Foundation (to YYasukochi), a Kurata grant awarded by the Hitachi Global Foundation (grant no. 1323 to YYasukochi and YYamada), the Core Research for Evolutional Science and Technology of the Japan Science and Technology Agency (grant no. JPMRJCR1302 to YYamada, JS and IT), and by KAKENHI grants from the Japan Society for the Promotion of Science (grant no. JP17H00758 to IT and YYasukochi; grant no. JP15H04772 to YYamada).

 Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request; if justified, applicable parts of the data can be made available in anonymized format.

 Authors' contributions

YYasukochi contributed to analysis and interpretation of the data, and to drafting of the manuscript. JS and IT contributed to analysis and interpretation of the data as well as revision of the manuscript. KK, MO, TF and HH each contributed to acquisition of the data and revision of the manuscript. YYamada contributed to conception and design of the study, and to acquisition, analysis and interpretation of the data, and revision of the manuscript. All authors read and approved the final manuscript.

 Ethics approval and consent to participate

All procedures performed in studies involving human participants were in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study protocol was approved by the Committees on the Ethics of Human Research of Mie University Graduate School of Medicine and Inabe General Hospital (Mie, Japan). Informed consent was obtained from all individual participants included in the study.

 Patient consent for publication

All participants provided written informed consent permitting publication of relevant data following anonymization of personal information.

 Competing interests

The authors declare that they have no competing interests.

 References GBD 2013 Mortality and Causes of Death CollaboratorsGlobal, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: A systematic analysis for the Global Burden of Disease Study 2013Lancet385117171201510.1016/S0140-6736(14)61682-225530442 BenjaminEJBlahaMJChiuveSECushmanMDasSRDeoRde FerrantiSDFloydJFornageMGillespieCAmerican Heart Association Statistics Committee and Stroke Statistics Subcommittee: Heart disease and stroke statistics-2017 update: A report from the American Heart AssociationCirculation135e146e603201710.1161/CIR.000000000000048528122885 National Center for Health StatisticsMortality multiple cause micro-data files, 2014: Public-use data file and documentation: NHLBI tabulations Ministry of Health, Labour and Welfare, JapanVital statistics SchunkertHKönigIRKathiresanSReillyMPAssimesTLHolmHPreussMStewartAFRBarbalicMGiegerCCardiogenics; CARDIoGRAM Consortium: Large-scale association analysis identifies 13 new susceptibility loci for coronary artery diseaseNat Genet43333338201110.1038/ng.78421378990 GolbusJRStitzielNOZhaoWXueCFarrallMMcPhersonRErdmannJDeloukasPWatkinsHSchunkertHCARDIoGRAMplusC4D, Myocardial Infarction Genetics (MIGen), Exome Sequencing Project and Early-Onset Myocardial Infarction (ESP EOMI), and the Pakistan Risk of Myocardial Infarction Study (PROMIS) Consortia*: Common and rare genetic variation in CCR2, CCR5, or CX3CR1 and risk of atherosclerotic coronary heart disease and glucometabolic traitsCirc Cardiovasc Genet9250258201610.1161/CIRCGENETICS.115.00137427013693 LuXWangLChenSHeLYangXShiYChengJZhangLGuCCHuangJCoronary ARtery DIsease Genome-Wide Replication And Meta-Analysis (CARDIoGRAM) Consortium: Genome-wide association study in Han Chinese identifies four new susceptibility loci for coronary artery diseaseNat Genet44890894201210.1038/ng.233722751097 NikpayMGoelAWonH-HHallLMWillenborgCKanoniSSaleheenDKyriakouTNelsonCPHopewellJCA comprehensive 1,000 Genomes-based genome-wide association meta-analysis of coronary artery diseaseNat Genet4711211130201510.1038/ng.339626343387 WakilSMRamRMuiyaNPMehtaMAndresEMazharNBazBHagosSAlshahidMMeyerBFA genome-wide association study reveals susceptibility loci for myocardial infarction/coronary artery disease in Saudi ArabsAtherosclerosisσ2456270201610.1016/j.atherosclerosis.2015.11.019 YamadaYFukuNTanakaMAoyagiYSawabeMMetokiNYoshidaHSatohKKatoKWatanabeSIdentification of CELSR1 as a susceptibility gene for ischemic stroke in Japanese individuals by a genome-wide association studyAtherosclerosis207144149200910.1016/j.atherosclerosis.2009.03.03819403135 KuboMHataJNinomiyaTMatsudaKYonemotoKNakanoTMatsushitaTYamazakiKOhnishiYSaitoSA nonsynonymous SNP in PRKCH (protein kinase C η) increases the risk of cerebral infarctionNat Genet39212217200710.1038/ng194517206144 GouveiaLOSobralJVicenteAMFerroJMOliveiraSAReplication of the CELSR1 association with ischemic stroke in a Portuguese case-control cohortAtherosclerosis217260262201110.1016/j.atherosclerosis.2011.03.02221511255 AkinyemiROOvbiageleBAkpaluAJenkinsCSagoeKOwolabiLSarfoFObiakoRGebreziabherMMelikamESIREN Investigators as Members of the H3Africa Consortium: Stroke genomics in people of African ancestry: Charting new pathsCardiovasc J Afr26Suppl 1S39S49201510.5830/CVJA-2015-03925962947 MeschiaJFNallsMMatarinMBrottTGBrownRDJrHardyJKisselaBRichSSSingletonAHernandezDSiblings With Ischemic Stroke Study Investigators: Siblings with ischemic stroke study: Results of a genome-wide scan for stroke lociStroke4227262732201110.1161/STROKEAHA.111.62048421940970 KilarskiLLAchterbergSDevanWJTraylorMMalikRLindgrenAPareGSharmaPSlowikAThijsVGARNET Collaborative Research Group, Wellcome Trust Case Control Consortium 2, Australian Stroke Genetic Collaborative, the METASTROKE Consortium, and the International Stroke Genetics Consortium: Meta-analysis in more than 17,900 cases of ischemic stroke reveals a novel association at 12q24.12Neurology83678685201410.1212/WNL.000000000000070725031287 YamadaYMatsuiKTakeuchiIFujimakiTAssociation of genetic variants with dyslipidemia and chronic kidney disease in a longitudinal population-based genetic epidemiological studyInt J Mol Med3512901300201510.3892/ijmm.2015.215125813695 YamadaYMatsuiKTakeuchiIOguriMFujimakiTAssociation of genetic variants of the α-kinase 1 gene with type 2 diabetes mellitus in a longitudinal population-based genetic epidemiological studyBiomed Rep3347354201510.3892/br.2015.43926137234 YamadaYMatsuiKTakeuchiIFujimakiTAssociation of genetic variants with coronary artery disease and ischemic stroke in a longitudinal population-based genetic epidemiological studyBiomed Rep3413419201510.3892/br.2015.44026137247 YamadaYMatsuiKTakeuchiIOguriMFujimakiTAssociation of genetic variants with hypertension in a longitudinal population-based genetic epidemiological studyInt J Mol Med3511891198201510.3892/ijmm.2015.215125813534 OguriMFujimakiTHoribeHKatoKMatsuiKTakeuchiIYamadaYObesity-related changes in clinical parameters and conditions in a longitudinal population-based epidemiological studyObes Res Clin Pract11299314201710.1016/j.orcp.2016.08.00827600639 GroveMLYuBCochranBJHarituniansTBisJCTaylorKDHansenMBoreckiIBCupplesLAFornageMBest practices and joint calling of the HumanExome BeadChip: The CHARGE ConsortiumPLoS One8e68095201310.1371/journal.pone.006809523874508 PriceALPattersonNJPlengeRMWeinblattMEShadickNAReichDPrincipal components analysis corrects for stratification in genome-wide association studiesNat Genet38904909200610.1038/ng184716862161 R Development Core Team: RA language and environment for statistical computingVienna, Austria2016 RStudio TeamRStudio: Integrated development environment for R. RStudio, Inc.Boston, MA2015 LiangKYZegerSLLongitudinal data analysis using generalized linear modelsBiometrika731322198610.1093/biomet/73.1.13 HanleyJANegassaAEdwardesMDForresterJEStatistical analysis of correlated data using generalized estimating equations: An orientationAm J Epidemiol157364375200310.1093/aje/kwf21512578807 HalekohUHøjsgaardSYanJThe R package geepack for generalized estimating equationsJ Stat Softw15111200610.18637/jss.v015.i02 WilliamsFMKCarterAMHysiPGSurdulescuGHodgkissDSoranzoNTraylorMBevanSDichgansMRothwellPMWEuroCLOT Investigators; Wellcome Trust Case Control Consortium 2; MOnica Risk, Genetics, Archiving and Monograph; MetaStroke; International Stroke Genetics Consortium: Ischemic stroke is associated with the ABO locus: The EuroCLOT studyAnn Neurol731631201310.1002/ana.2383823381943 HolmenOLZhangHFanYHovelsonDHSchmidtEMZhouWGuoYZhangJLanghammerALøchenMLSystematic evaluation of coding variation identifies a candidate causal variant in TM6SF2 influencing total cholesterol and myocardial infarction riskNat Genet46345351201410.1038/ng.292624633158 BenjaminiYHochbergYControlling the false discovery rate: A practical and powerful approach to multiple testingJ R Stat Soc B572893001995 SitlaniCMRiceKMLumleyTMcKnightBCupplesLAAveryCLNoordamRStrickerBHCWhitselEAPsatyBMGeneralized estimating equations for genome-wide association studies using longitudinal phenotype dataStat Med34118130201510.1002/sim.632325297442 VoormanARiceKLumleyTFast computation for genome-wide association studies using boosted one-step statisticsBioinformatics2818181822201210.1093/bioinformatics/bts29122592383 1000 Genomes Project ConsortiumAbecasisGRAltshulerDAutonABrooksLDDurbinRMGibbsRAHurlesMEMcVeanGAA map of human genome variation from population-scale sequencingNature46710611073201010.1038/nature0953420981092 MachielaMJChanockSJLDlink: A web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variantsBioinformatics3135553557201510.1093/bioinformatics/btv40226139635 LeslieRO'DonnellCJJohnsonADGRASP: Analysis of genotype-phenotype results from 1390 genome-wide association studies and corresponding open access databaseBioinformatics30i185i194201410.1093/bioinformatics/btu27324931982 PiñeroJQueralt-RosinachNBravoÀDeu-PonsJBauer-MehrenABaronMSanzFFurlongLIDisGeNET: A discovery platform for the dynamical exploration of human diseases and their genes. Database (Oxford)2015bav028201525877637 MacArthurJBowlerECerezoMGilLHallPHastingsEJunkinsHMcMahonAMilanoAMoralesJThe new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog)Nucleic Acids Res45D1D896D901201710.1093/nar/gkw113327899670 Warde-FarleyDDonaldsonSLComesOZuberiKBadrawiRChaoPFranzMGrouiosCKaziFLopesCTThe GeneMANIA prediction server: Biological network integration for gene prioritization and predicting gene functionNucleic Acids Res38W214220201010.1093/nar/gkq53720576703 MontojoJZuberiKRodriguezHBaderGDMorrisQGeneMANIA: Fast gene network construction and function prediction for CytoscapeF1000Res3153201425254104 MontojoJZuberiKRodriguezHKaziFWrightGDonaldsonSLMorrisQBaderGDGeneMANIA Cytoscape plugin: Fast gene function predictions on the desktopBioinformatics2629272928201010.1093/bioinformatics/btq56220926419 ShannonPMarkielAOzierOBaligaNSWangJTRamageDAminNSchwikowskiBIdekerTCytoscape: A software environment for integrated models of biomolecular interaction networksGenome Res1324982504200310.1101/gr.123930314597658 SherrySTWardMHKholodovMBakerJPhanLSmigielskiEMSirotkinKdbSNP: The NCBI database of genetic variationNucleic Acids Res29308311200110.1093/nar/29.1.30811125122 WarnesGGorjancGLeischFManMGenetics: Population Genetics2013 StaceyMLinHHHilyardKLGordonSMcKnightAJHuman epidermal growth factor (EGF) module-containing mucin-like hormone receptor 3 is a new member of the EGF-TM7 family that recognizes a ligand on human macrophages and activated neutrophilsJ Biol Chem2761886318870200110.1074/jbc.M10114720011279179 LibbyPInflammation in atherosclerosisNature420868874200210.1038/nature0132312490960 GreavesDRGordonSImmunity, atherosclerosis and cardiovascular diseaseTrends Immunol22180181200110.1016/S1471-4906(00)01848-211394351 LibbyPInflammation in atherosclerosisArterioscler Thromb Vasc Biol3220452051201210.1161/ATVBAHA.108.17970522895665 UhlénMFagerbergLHallströmBMLindskogCOksvoldPMardinogluASivertssonÅKampfCSjöstedtEAsplundAProteomicsTissue-based map of the human proteome. Science34712604192015 ThulPJÅkessonLWikingMMahdessianDGeladakiAAit BlalHAlmTAsplundABjörkLBreckelsLMA subcellular map of the human proteomeScience356eaal3321201710.1126/science.aal332128495876 WangCLiuXLiuYZhangQYaoZHuangBZhangPLiNCaoXZinc finger protein 64 promotes Toll-like receptor-triggered proinflammatory and type I interferon production in macrophages by enhancing p65 subunit activationJ Biol Chem2882460024608201310.1074/jbc.M113.47339723857586 SomiaNVSchmittMJVetterDEvan AntwerpDHeinemannSFVermaIMLFG: An anti-apoptotic gene that provides protection from Fas-mediated cell deathProc Natl Acad Sci USA961266712672199910.1073/pnas.96.22.1266710535980 StrasserAJostPJNagataSThe many roles of FAS receptor signaling in the immune systemImmunity30180192200910.1016/j.immuni.2009.01.00119239902 SchützCOelkeMSchneckJPMackensenAFleckMKiller artificial antigen-presenting cells: The synthetic embodiment of a ‘guided missile’Immunotherapy2539550201010.2217/imt.10.2620636007 LenardoMJFas and the art of lymphocyte maintenanceJ Exp Med183721724199610.1084/jem.183.3.7218642275 HanadaRLeibbrandtAHanadaTKitaokaSFuruyashikiTFujiharaHTrichereauJPaolinoMQadriFPlehmRCentral control of fever and female body temperature by RANKL/RANKNature462505509200910.1038/nature0859619940926 ShimamuraMNakagamiHOsakoMKKurinamiHKoriyamaHZhengdaPTomiokaHTenmaAWakayamaKMorishitaROPG/RANKL/RANK axis is a critical inflammatory signaling system in ischemic brain in miceProc Natl Acad Sci USA11181918196201410.1073/pnas.140054411124847069

Diagram of GeneMANIA network for the presently identified genes (closed red circles) and previously reported genes associated with CAD (closed blue circles) and CI (closed green circles) or both (closed light blue circles). The phenotypes (CAD or CI) associated with the identified genes are shown in parentheses. CAD, coronary artery disease. CI, cerebral infarction.

Longitudinal characteristics of study subjects in the Inabe cohort.

Characteristic Controla CADa CIa
No. of subjectsb 5,713 170 117
Sex, male/female, %b 54.7/45.3 77.1/22.9 68.4/31.6
Age, years 52.0±0.07 (27,143) 63.1±0.41 (619) 63.4±0.35 (466)
Body mass index, kg/m2 22.9±0.02 (26,719) 23.9±0.13 (507) 23.6±0.16 (416)
Current or former smoker, % 38.4 (27,143) 40.1 (619) 35.0 (466)
Hypertension, % 31.9 (26,715) 75.7 (503) 65.1 (415)
Systolic blood pressure, mmHg 120.4±0.10 (26,715) 126.5±0.77 (503) 127.6±0.81 (415)
Diastolic blood pressure, mmHg 74.6±0.08 (26,715) 76.5±0.55 (503) 77.5±0.58 (415)
Type 2 diabetes mellitus, % 11.4 (26,745) 38.8 (590) 36.4 (464)
Fasting plasma glucose, mmol/l 5.59±0.007 (26,731) 6.49±0.086 (586) 6.22±0.098 (464)
Blood hemoglobin A1c, % 5.69±0.004 (19,870) 6.18±0.052 (465) 6.03±0.041 (410)
Dyslipidemia, % 58.4 (26,713) 76.5 (592) 77.9 (444)
Serum triglycerides, mmol/l 1.25±0.006 (26,710) 1.48±0.040 (591) 1.36±0.037 (444)
Serum HDL-cholesterol, mmol/l 1.61±0.003 (26,687) 1.43±0.019 (585) 1.51±0.020 (440)
Serum LDL-cholesterol, mmol/l 3.19±0.005 (25,594) 2.76±0.032 (563) 3.20±0.039 (422)
Chronic kidney disease, % 10.8 (24,461) 37.6 (582) 37.0 (441)
Serum creatinine, µmol/l 74.1±0.58 (24,461) 132.7±8.65 (582) 163.0±13.33 (441)
eGFR, ml min−1 1.73 m−2 79.2±0.11 (24,461) 66.1±1.11 (582) 64.5±1.31 (441)
Hyperuricemia, % 18.1 (23,985) 42.3 (575) 27.5 (436)
Serum uric acid, µmol/l 327.7±0.56 (23,985) 367.3±4.20 (575) 352.7±4.09 (436)

Values in parentheses indicate the number of measurements taken

numbers based on examination data from the final visit for each subject. Quantitative data are presented as the mean and standard error of the mean. CAD, coronary artery disease; CI, cerebral infarction; HDL, high-density lipoprotein; LDL, low-density lipoprotein; eGFR, estimated glomerular filtration rate.

SNPs significantly (FDR <0.05, approxdf >30) associated with CAD or CI.

Disease (genetic model) RefSNP ID Nucleotide (amino acid) substitutiona Positionb Gene Estimatec Std. err. P-value MAF Apxdfd FDR
CAD (additive)e rs4606855 G→C (E75Q) 19: 14,658,527 ADGRE3 −0.014 0.003 2.5×10−6 0.120 71 0.031
rs3746414 G→A (S451N) 20: 52,152,840 ZFP64 −0.013 0.003 5.9×10−6 0.140 93 0.048
CAD (recessive)f rs7132908 G→A 12: 49,869,365 FAIM2 −8.9×1014 2.2×1013 <2.0×10−16 0.291 65 <4.9×10−12
CI (recessive)f rs6580741 G→C (H2228Q) 12: 50,333,923 FAM186A −2.2×1015 2.2×1013 <2.0×10−16 0.260 48 <4.9×10−12
rs1324015 G→A 13: 43,153,713 LINC00400 −40.22 0.144 <2.0×10−16 0.264 49 <4.9×10−12
rs884205 G→T 18: 62,387,624 TNFRSF11A −40.12 0.147 <2.0×10−16 0.220 32 <4.9×10−12

Major allele → Minor allele

position in NCBI Genome Reference Consortium Human Build 38

estimate of correlation coefficient

approxdf = 2 × MAF × Nindep, where Nindep = the sum of the estimated number of independent observations per person

additive model [AA vs. AB vs. BB (A, major allele; B, minor allele)]

recessive model (AA + AB vs. BB). SNP, single nucleotide polymorphism; FDR, false discovery rate; CAD, coronary artery disease; CI, cerebral infarction; MAF, minor allele frequency; Std. err., standard error.

Genotype distributions for three candidate SNPs among subjects with CAD and controls.

RefSNP ID Positiona Gene Genotype Controlsb CADb
rs4606855 19: 14,658,527 ADGRE3 GG 21,095 (77.7) 564 (91.1)
GC 5,676 (20.9) 54 (8.7)
CC 372 (1.4) 1 (0.2)
rs3746414 20: 52,152,840 ZFP64 GG 20,122 (74.1) 529 (85.5)
GA 6,506 (24.0) 81 (13.1)
AA 515 (1.9) 9 (1.5)
rs7132908 12: 49,869,365 FAIM2 GG 13,609 (50.1) 344 (55.6)
GA 11,128 (41.0) 252 (40.7)
AA 2,406 (8.9) 23 (3.7)

Position in NCBI Genome Reference Consortium Human Build 38

values indicate the numbers of measurements taken, with the percentages in parentheses. SNP, single nucleotide polymorphism; CAD, coronary artery disease.

Genotype distributions for three candidate SNPs among subjects with CI and controls.

RefSNP ID Positiona Gene Genotype Controlsb CIb
rs6580741 12: 50,333,923 FAM186A GG 14,669 (54.1) 269 (57.7)
GC 10,555 (38.9) 171 (36.7)
CC 1,900 (7.0) 26 (5.6)
rs1324015 13: 43,153,713 LINC00400 GG 14,809 (54.6) 295 (63.3)
GA 10,433 (38.5) 171 (36.7)
AA 1,882 (6.9) 0 (0.0)
rs884205 18: 62,387,624 TNFRSF11A GG 16,573 (61.1) 312 (67.0)
GT 9,154 (33.7) 154 (33.0)
TT 1,397 (5.2) 0 (0.0)

Position in NCBI Genome Reference Consortium Human Build 38

values indicate the numbers of measurements taken, with the percentages in parentheses. SNP, single nucleotide polymorphism; CI, cerebral infarction.

Correlation estimates of candidate (query) SNPs with adjacent SNPs using LDproxy.

Chr. pos. RefSNP ID r2 Chr. pos. RefSNP ID r2 Chr. pos. RefSNP ID r2
18q21.33 rs884205 1.00 20q13.2 rs12481281 0.77 12q13.12 rs7979830 1.00
13q14.11 rs1324015 1.00 20q13.2 rs3787176 0.77 12q13.12 rs35723625 1.00
13q14.11 rs2589313 0.59 20q13.2 rs72626582 0.77 12q13.12 rs10876013 1.00
13q14.11 rs1324016 0.57 20q13.2 rs72626583 0.77 12q13.12 rs12425229 1.00
13q14.11 rs2762188 0.57 20q13.2 rs6021711 0.76 12q13.12 rs3812825 1.00
13q14.11 rs2589314 0.57 20q13.2 rs2180366 0.76 12q13.12 rs1362983 1.00
13q14.11 rs35574382 0.57 20q13.2 rs67904269 0.76 12q13.12 rs141978158 1.00
13q14.11 rs2762185 0.57 20q13.2 rs7273288 0.74 12q13.12 rs7308095 1.00
13q14.11 rs2657098 0.57 20q13.2 rs13045200 0.61 12q13.12 rs11169319 1.00
13q14.11 rs1562123 0.57 20q13.2 rs11478756 0.61 12q13.12 rs12811291 1.00
13q14.11 rs11424017 0.52 20q13.2 rs11473560 0.56 12q13.12 rs34825838 1.00
20q13.2 rs3746414 1.00 20q13.2 rs6021739 0.56 12q13.12 rs11169317 1.00
20q13.2 rs3746413 1.00 20q13.2 rs6096795 0.56 12q13.12 rs3861100 1.00
20q13.2 rs3746415 1.00 20q13.2 rs6021738 0.56 12q13.12 rs11169315 1.00
20q13.2 rs3787181 1.00 20q13.2 rs6021737 0.56 12q13.12 rs2009072 1.00
20q13.2 rs11431529 1.00 20q13.2 rs6021736 0.56 12q13.12 rs11169314 1.00
20q13.2 rs67642347 1.00 20q13.2 rs6013400 0.56 12q13.12 rs6580739 0.97
20q13.2 rs4809893 1.00 20q13.2 rs6096794 0.56 12q13.12 rs7972824 0.97
20q13.2 rs8122116 1.00 20q13.2 rs6013399 0.56 12q13.12 rs9739363 0.97
20q13.2 rs6021732 1.00 20q13.2 rs117265881 0.56 12q13.12 rs12823506 0.97
20q13.2 rs2224227 1.00 20q13.2 rs4809892 0.51 12q13.12 rs112456855 0.97
20q13.2 rs66953446 1.00 20q13.2 rs145753021 0.51 12q13.12 rs12422417 0.97
20q13.2 rs115523244 1.00 19p13.12 rs4606855 1.00 12q13.12 rs35224873 0.97
20q13.2 rs6021723 1.00 19p13.12 rs56819877 1.00 12q13.12 rs12828340 0.87
20q13.2 rs12480406 1.00 19p13.12 rs4808971 1.00 12q13.12 rs17124562 0.87
20q13.2 rs67638808 1.00 19p13.12 rs7255012 1.00 12q13.12 rs11169335 0.87
20q13.2 rs67262858 1.00 19p13.12 rs10426121 1.00 12q13.12 rs35576436 0.87
20q13.2 rs6013397 1.00 19p13.12 rs6511956 1.00 12q13.12 rs11169332 0.87
20q13.2 rs12481222 0.97 19p13.12 rs4273164 1.00 12q13.12 rs11169331 0.87
20q13.2 rs139218250 0.97 19p13.12 rs4239642 1.00 12q13.12 rs71465002 0.87
20q13.2 rs7346642 0.97 19p13.12 rs10402993 0.95 12q13.12 rs7314465 0.87
20q13.2 rs67892028 0.97 19p13.12 rs11085902 0.95 12q13.12 rs6580730 0.87
20q13.2 rs73130324 0.97 19p13.12 rs11085901 0.95 12q13.12 rs7308885 0.87
20q13.2 rs75085690 0.97 19p13.12 rs7245656 0.86 12q13.12 rs2358539 0.87
20q13.2 rs3838014 0.94 19p13.12 rs7250114 0.86 12q13.12 rs201343445 0.87
20q13.2 rs3179313 0.94 19p13.12 rs73506161 0.86 12q13.12 rs12424713 0.87
20q13.2 rs3179314 0.94 19p13.12 rs35805282 0.86 12q13.12 rs12424691 0.87
20q13.2 rs3787178 0.94 19p13.12 rs75540045 0.86 12q13.12 rs11169322 0.87
20q13.2 rs1555328 0.94 19p13.12 rs34930135 0.86 12q13.12 rs12425705 0.87
20q13.2 rs6021748 0.94 19p13.12 rs373306807 0.66 12q13.12 rs35628283 0.87
20q13.2 rs6021747 0.94 12q13.12 rs6580741 1.00 12q13.12 rs35875720 0.87
20q13.2 rs4811297 0.94 12q13.12 rs7310541 1.00 12q13.12 rs34198664 0.84
20q13.2 rs111897744 0.94 12q13.12 rs7134337 1.00 12q13.12 rs11292692 0.82
20q13.2 rs6021745 0.94 12q13.12 rs7134595 1.00 12q13.12 rs4768905 0.77
20q13.2 rs6021744 0.94 12q13.12 rs55931113 1.00 12q13.12 rs59210472 0.77
20q13.2 rs6021743 0.94 12q13.12 rs35878271 1.00 12q13.12 rs4636745 0.77
20q13.2 rs6021742 0.94 12q13.12 rs34039674 1.00 12q13.12 rs11830586 0.77
20q13.2 rs6021741 0.94 12q13.12 rs66895907 1.00 12q13.12 rs7973910 0.77
20q13.2 rs6021740 0.94 12q13.12 rs4768900 1.00 12q13.12 rs73305103 0.77
20q13.2 rs6021735 0.94 12q13.12 rs10876020 1.00 12q13.12 rs4321029 0.77
20q13.2 rs6021734 0.94 12q13.12 rs34858415 1.00 12q13.12 rs7968898 0.77
20q13.2 rs56783695 0.94 12q13.12 rs11836169 1.00 12q13.12 rs151015253 0.77
20q13.2 rs60610902 0.94 12q13.12 rs4768949 1.00 12q13.12 rs11169323 0.74
20q13.2 rs6021733 0.94 12q13.12 rs4768872 1.00 12q13.12 rs73305105 0.74
20q13.2 rs60614585 0.94 12q13.12 rs34614542 1.00 12q13.12 rs35925338 0.74
20q13.2 rs57913961 0.94 12q13.12 rs12312177 1.00 12q13.12 NA 0.72
20q13.2 rs6021730 0.94 12q13.12 rs34894919 1.00 12q13.12 rs144615146 0.72
20q13.2 rs6021729 0.94 12q13.12 rs12424876 1.00 12q13.12 rs78601155 0.72
20q13.2 rs6021728 0.94 12q13.12 rs11169377 1.00 12q13.12 rs74336127 0.72
20q13.2 rs6021727 0.94 12q13.12 rs11169376 1.00 12q13.12 rs141852922 0.69
20q13.2 rs6096791 0.94 12q13.12 rs36017775 1.00 12q13.12 rs145504356 0.69
20q13.2 rs6091458 0.94 12q13.12 rs11169375 1.00 12q13.12 rs34098872 0.65
20q13.2 rs6091457 0.94 12q13.12 rs7137319 1.00 12q13.12 rs67576611 0.64
20q13.2 rs200887094 0.94 12q13.12 rs7295847 1.00 12q13.12 rs7138622 0.64
20q13.2 rs6096784 0.94 12q13.12 rs7296291 1.00 12q13.12 rs7138420 0.64
20q13.2 rs77290230 0.94 12q13.12 rs7312252 1.00 12q13.12 rs7315690 0.64
20q13.2 rs12624632 0.94 12q13.12 rs10506292 1.00 12q13.12 rs5798135 0.63
20q13.2 rs6021722 0.94 12q13.12 rs34849043 1.00 12q13.12 rs17124514 0.62
20q13.2 rs6091453 0.94 12q13.12 rs10615610 1.00 12q13.12 rs12424335 0.62
20q13.2 rs6021721 0.94 12q13.12 rs10876023 1.00 12q13.12 rs34145380 0.62
20q13.2 rs6021720 0.94 12q13.12 rs10876024 1.00 12q13.12 rs34309034 0.62
20q13.2 rs6512796 0.94 12q13.12 rs7301186 1.00 12q13.12 rs34245511 0.62
20q13.2 rs12480343 0.93 12q13.12 rs11169374 1.00 12q13.12 rs9364 0.62
20q13.2 rs12480321 0.93 12q13.12 rs11169373 1.00 12q13.12 rs8181651 0.60
20q13.2 rs67451131 0.93 12q13.12 rs35209607 1.00 12q13.12 rs80003859 0.60
20q13.2 rs6013393 0.93 12q13.12 rs4421818 1.00 12q13.12 rs144939089 0.60
20q13.2 rs67862732 0.93 12q13.12 rs11169370 1.00 12q13.12 rs28364704 0.59
20q13.2 rs144698812 0.93 12q13.12 rs11169369 1.00 12q13.12 rs736167 0.59
20q13.2 rs117427009 0.93 12q13.12 rs11169390 1.00 12q13.12 rs10783347 0.59
20q13.2 rs140670817 0.93 12q13.12 rs11169391 1.00 12q13.12 rs10783346 0.59
20q13.2 rs6021709 0.93 12q13.12 rs4445717 1.00 12q13.12 rs61928279 0.59
20q13.2 rs6021708 0.93 12q13.12 rs9668187 1.00 12q13.12 rs7302422 0.59
20q13.2 rs72626578 0.93 12q13.12 rs60025018 1.00 12q13.12 rs12423130 0.59
20q13.2 rs6021707 0.93 12q13.12 rs11169393 1.00 12q13.12 rs3815671 0.59
20q13.2 rs6021706 0.93 12q13.12 rs10467106 1.00 12q13.12 rs57583527 0.59
20q13.2 rs140980517 0.93 12q13.12 rs12303082 1.00 12q13.12 rs10747572 0.59
20q13.2 rs143481833 0.93 12q13.12 rs7971374 1.00 12q13.12 rs71083515 0.57
20q13.2 rs12624823 0.93 12q13.12 rs10783352 1.00 12q13.12 rs79043170 0.55
20q13.2 rs12624819 0.93 12q13.12 rs10783353 1.00 12q13.12 rs35404088 0.55
20q13.2 rs148218979 0.93 12q13.12 rs12299758 1.00 12q13.12 rs10783340 0.55
20q13.2 rs6021760 0.90 12q13.12 rs12299669 1.00 12q13.12 rs7978904 0.55
20q13.2 rs3818198 0.90 12q13.12 rs11169394 1.00 12q13.12 rs10783338 0.55
20q13.2 rs72626580 0.90 12q13.12 rs11169395 1.00 12q13.12 rs7398567 0.55
20q13.2 rs4811301 0.90 12q13.12 rs11169367 1.00 12q13.12 rs10735824 0.55
20q13.2 rs4811303 0.90 12q13.12 rs11833608 1.00 12q13.12 rs7972465 0.55
20q13.2 rs4811304 0.90 12q13.12 rs113486728 1.00 12q13.12 rs6580728 0.55
20q13.2 rs4811305 0.90 12q13.12 rs376666931 1.00 12q13.12 rs7138945 0.55
20q13.2 rs140407501 0.90 12q13.12 rs7980911 1.00 12q13.12 rs11169282 0.55
20q13.2 rs6021763 0.90 12q13.12 rs7302363 1.00 12q13.12 rs57061317 0.53
20q13.2 rs6021764 0.90 12q13.12 rs4768855 1.00 12q13.12 rs76382737 0.53
20q13.2 rs6021765 0.90 12q13.12 rs6580737 1.00 12q13.12 rs12369104 0.53
20q13.2 rs6021766 0.90 12q13.12 rs7488682 1.00 12q13.12 rs12819883 0.53
20q13.2 rs11480360 0.90 12q13.12 rs7972068 1.00 12q13.12 rs3741562 0.53
20q13.2 rs6021767 0.90 12q13.12 rs7972202 1.00 12q13.12 rs113411336 0.53
20q13.2 rs6021768 0.90 12q13.12 rs6580743 1.00 12q13.12 rs34632215 0.53
20q13.2 rs6013409 0.90 12q13.12 rs4348979 1.00 12q13.12 rs11169278 0.53
20q13.2 rs6021770 0.90 12q13.12 rs11169360 1.00 12q13.12 rs1554845 0.53
20q13.2 rs2038429 0.90 12q13.12 rs11169359 1.00 12q13.12 rs2204684 0.53
20q13.2 rs6021772 0.90 12q13.12 rs12814094 1.00 12q13.12 rs2204683 0.53
20q13.2 rs2273472 0.90 12q13.12 rs7315955 1.00 12q13.12 rs10876000 0.53
20q13.2 rs2038430 0.90 12q13.12 rs11169357 1.00 12q13.12 rs7967979 0.53
20q13.2 rs6021774 0.90 12q13.12 rs10876028 1.00 12q13.12 rs34167640 0.53
20q13.2 rs58878184 0.90 12q13.12 rs7135777 1.00 12q13.12 rs7307469 0.53
20q13.2 rs72626581 0.90 12q13.12 rs11838347 1.00 12q13.12 rs1554844 0.53
20q13.2 rs34963386 0.90 12q13.12 rs12426444 1.00 12q13.12 rs7961065 0.53
20q13.2 NA 0.90 12q13.12 rs9705460 1.00 12q13.12 rs7961112 0.53
20q13.2 rs6021775 0.90 12q13.12 rs7304445 1.00 12q13.12 rs7294618 0.53
20q13.2 rs1973951 0.90 12q13.12 rs11323536 1.00 12q13.12 rs7307230 0.53
20q13.2 rs77286509 0.90 12q13.12 rs7311973 1.00 12q13.12 rs7135322 0.53
20q13.2 rs7265436 0.90 12q13.12 rs10876017 1.00 12q13.12 rs59262224 0.53
20q13.2 rs6096760 0.90 12q13.12 rs10876016 1.00 12q13.12 rs17124432 0.53
20q13.2 rs142762031 0.87 12q13.12 rs7974648 1.00 12q13.12 rs12369049 0.53
20q13.2 rs138459074 0.87 12q13.12 rs12821454 1.00 12q13.12 rs7308474 0.53
20q13.2 rs67243058 0.84 12q13.12 rs7305995 1.00 12q13.12 rs7968119 0.53
20q13.2 rs6021731 0.82 12q13.12 rs35933908 1.00 12q13.12 rs61529910 0.51
20q13.2 rs6013396 0.79 12q13.12 rs12832940 1.00 12q13.12 NA 0.50
20q13.2 rs6021719 0.79 12q13.12 rs10876015 1.00 12q13.12 rs7132908 1.00
20q13.2 rs6021718 0.79 12q13.12 rs10876014 1.00 12q13.12 rs3205718 1.00
20q13.2 rs6091452 0.79 12q13.12 rs67138019 1.00 12q13.12 rs12146733 0.98
20q13.2 rs6091451 0.79 12q13.12 rs11169351 1.00 12q13.12 rs1893492 0.98
20q13.2 rs6126474 0.79 12q13.12 rs11169350 1.00 12q13.12 rs145103902 0.98
20q13.2 rs6123134 0.79 12q13.12 rs11169349 1.00 12q13.12 rs73116325 0.98
20q13.2 rs8121769 0.79 12q13.12 rs35535298 1.00 12q13.12 rs146875448 0.85
20q13.2 rs6021716 0.79 12q13.12 rs2111988 1.00 12q13.12 rs11169199 0.80
20q13.2 rs6021715 0.79 12q13.12 rs7311491 1.00 12q13.12 rs297924 0.80
20q13.2 rs6123133 0.79 12q13.12 rs7311378 1.00 12q13.12 rs17201502 0.80
20q13.2 rs6096776 0.79 12q13.12 rs11169348 1.00 12q13.12 rs73116335 0.80
20q13.2 rs62216897 0.79 12q13.12 rs6580735 1.00 12q13.12 rs73116338 0.80
20q13.2 rs6126472 0.79 12q13.12 rs11169347 1.00 12q13.12 rs73116339 0.80
20q13.2 rs7265434 0.79 12q13.12 rs11169345 1.00 12q13.12 rs145512623 0.75
20q13.2 rs7273176 0.79 12q13.12 rs7956468 1.00 12q13.12 NA 0.59
20q13.2 rs7273121 0.79 12q13.12 rs1972611 1.00 12q13.12 rs149596227 0.57
20q13.2 rs8183138 0.79 12q13.12 rs10783344 1.00 12q13.12 rs373555454 0.55
20q13.2 rs6123131 0.79 12q13.12 rs35768991 1.00 12q13.12 rs12367809 0.53
20q13.2 rs79718879 0.79 12q13.12 rs11169339 1.00 12q13.12 rs7306275 0.51
20q13.2 rs76931275 0.79 12q13.12 rs7486747 1.00 12q13.12 rs7138803 0.50
20q13.2 rs6021714 0.79 12q13.12 rs7953953 1.00 12q13.12 rs112502508 0.50
20q13.2 rs6021713 0.79 12q13.12 rs7132551 1.00

SNPs with r2>0.5 only are shown. The candidate SNPs identified in the present study are shown in bold. SNP, single nucleotide polymorphism; Chr. pos., chromosomal position; NA, not available.