Recent genetic association studies have identified 55 genetic loci associated with obesity or body mass index (BMI). The vast majority, 51 loci, however, were identified in European-ancestry populations. We conducted a meta-analysis of associations between BMI and ∼2.5 million genotyped or imputed single nucleotide polymorphisms among 86 757 individuals of Asian ancestry, followed by in silico and de novo replication among 7488–47 352 additional Asian-ancestry individuals. We identified four novel BMI-associated loci near the KCNQ1 (rs2237892, P = 9.29 × 10−13), ALDH2/MYL2 (rs671, P = 3.40 × 10−11; rs12229654, P = 4.56 × 10−9), ITIH4 (rs2535633, P = 1.77 × 10−10) and NT5C2 (rs11191580, P = 3.83 × 10−8) genes. The association of BMI with rs2237892, rs671 and rs12229654 was significantly stronger among men than among women. Of the 51 BMI-associated loci initially identified in European-ancestry populations, we confirmed eight loci at the genome-wide significance level (P < 5.0 × 10−8) and an additional 14 at P < 1.0 × 10−3 with the same direction of effect as reported previously. Findings from this analysis expand our knowledge of the genetic basis of obesity.

INTRODUCTION

To date, genome-wide association studies (GWAS) have identified 55 genetic loci associated with obesity or body mass index (BMI) (114). Fifty-one of these loci were reported by studies conducted in populations of European ancestry. The remaining four loci were identified by our meta-analyses conducted among East Asians (9,10). However, these loci together explain only a small portion of observed variation in BMI [1.45% in Europeans (8), 1.18% in East Asians (9)], suggesting that additional BMI-related loci remain to be discovered. Since the publication of our previous meta-analysis in East Asians (9,10), nine additional GWAS with 18 352 additional participants have joined the Asian Genetic Epidemiology Network (AGEN) BMI-Consortium. We carried out a new round of meta-analyses that included data from 86 757 Asians recruited from 21 studies conducted in mainland China, Japan, Singapore, South Korea, Taiwan, the Philippines and the USA to identify new BMI loci and re-confirm associations with BMI that have been previously reported.

RESULTS

Our initial meta-analysis used BMI as the outcome and analyzed the association of BMI with ∼2.5 million genotyped or imputed single nucleotide polymorphisms (SNPs) generated from these 21 studies, comprising 86 757 individuals of East Asian or Southeast Asian ancestry (Stage I). This was followed by a replication analysis (Stage II) of eight selected SNPs from four study sites, comprising 7488–47 352 Asian-ancestry individuals based on the availability of de novo and/or in silico data for each SNP. Details of the study design are presented in Supplementary Material, Figure S1. Participating studies are described in the Supplementary Information and Supplementary Material, Tables S1–S3.

The Stage I meta-analysis found eight SNPs at seven loci near the KCNQ1 (rs2237892, P = 7.32 × 10−10), ALDH2/MYL2 (rs671, P = 5.96 × 10−10, rs12229654, P = 1.26 × 10−8), ITIH4 (rs2535633, P = 1.33 × 10−8), NT5C2 (rs11191580, P = 7.59 × 10−6), LINC00461 (rs6893807, P = 1.81 × 10−7) and SEMA6D (rs1912631, P = 6.06 × 10−8) genes and the intergenic region at 2p25.3 (rs4854307, P = 9.21 × 10−7) that were associated with BMI at or near the genome-wide significance level (Table 1, Supplementary Material, Table S4). These eight SNPs were taken forward to the Stage II replication analyses (Supplementary Material, Table S3), which included de novo genotyping data from three study sites with a total of 40 422 participants and in silico replication data from the Tai Chi study (N = 7369) genotyped with Illumina's iSelect 200 k Cardio-MetaboChip (Supplementary Material, Table S4). In the Stage II analysis, five of these eight SNPs had the same direction of association as in Stage I and were nominally significant (P < 0.05). Combined analysis of data from Stages I and II showed that the association for all five of these SNPs at four genetic loci reached the genome-wide significance level: KCNQ1 (rs2237892, P = 9.29 × 10−13), ALDH2/MYL2 (rs671, P = 3.40 × 10−11, rs12229654, P = 4.56 × 10−9), ITIH4 (rs2535633, P = 1.77 × 10−10) and NT5C2 (rs11191580, P = 3.83 × 10−8) (Table 1, Supplementary Material, Table S4). Data obtained from the GIANT consortium (8,15) (Supplementary Material, Table S5) revealed significant associations for two of the SNPs (P = 9.18 × 10−3 for rs2535633 and P = 1.06 × 10−8 for rs11191580) with the same direction of association as the current study. The SNPs in the ALDH2 (rs671) and MYL2 (rs12229654) genes had a minor allele frequency (MAF) of 0.24 and 0.20, respectively, in the current study, but are monomorphic in HapMap European-ancestry data; no GIANT consortium data were available for these two SNPs. The variation explained by each newly identified SNP ranged from 0.03% to 0.05% (Table 1, Supplementary Material, Table S4). The variation explained for all four of these newly identified BMI loci combined was 0.16% based on Stage II data.

Table 1.

Newly identified loci associated with BMI variation in Asian-ancestry populations

Nearby gene Cytoband SNP Allelesa EAFb Stage I P Stage I and II
 
EV (%)e 
      Stage II P Number of samples β (SE)c Pd  
KCNQ1 11p15.4 rs2237892 T/C 0.36 7.32E−10 1.73E−04 133 312 0.0298 (0.0042) 9.29E−13 0.04 
ALDH2 12q24.12 rs671 G/A 0.76 5.96E−10 6.64E−03 97 990 0.0378 (0.0057) 3.40E−11 0.05 
MYL2 12q24.11 rs12229654 T/G 0.80 1.26E−08 1.89E−02 110 211 0.0341 (0.0058) 4.56E−09 0.04 
ITIH4 3p21.1 rs2535633 G/C 0.42 1.33E−08 2.56E−03 111 673 0.0288 (0.0045) 1.77E−10 0.04 
NT5C2 10q24.33 rs11191580 C/T 0.27 7.59E−06 6.78E−04 98 883 0.0295 (0.0054) 3.83E−08 0.03 
Nearby gene Cytoband SNP Allelesa EAFb Stage I P Stage I and II
 
EV (%)e 
      Stage II P Number of samples β (SE)c Pd  
KCNQ1 11p15.4 rs2237892 T/C 0.36 7.32E−10 1.73E−04 133 312 0.0298 (0.0042) 9.29E−13 0.04 
ALDH2 12q24.12 rs671 G/A 0.76 5.96E−10 6.64E−03 97 990 0.0378 (0.0057) 3.40E−11 0.05 
MYL2 12q24.11 rs12229654 T/G 0.80 1.26E−08 1.89E−02 110 211 0.0341 (0.0058) 4.56E−09 0.04 
ITIH4 3p21.1 rs2535633 G/C 0.42 1.33E−08 2.56E−03 111 673 0.0288 (0.0045) 1.77E−10 0.04 
NT5C2 10q24.33 rs11191580 C/T 0.27 7.59E−06 6.78E−04 98 883 0.0295 (0.0054) 3.83E−08 0.03 

aShown as: effect allele/other allele.

bEffect allele frequency in Asian-ancestry populations, estimated from Stage I and II studies.

cPer allele effects of SNPs on BMI are presented in standard deviations, which were derived from the meta-analysis.

dDerived from the meta-analysis. The P-values for combined data were adjusted for both study-specific inflation factors and the estimated inflation factor for the Stage I meta-analysis statistic.

eExplained variance, estimated from combined Stages I and II data.

The two newly identified SNPs, rs671 in the ALDH2 gene (12q24.12) and rs12229654 in the MYL2 gene (12q24.11), are located 827 kb apart and are in LD (r2 = 0.58) in Asians (Fig. 1). To examine their independent effects, we conducted a conditional analysis that included these two SNPs in the same regression model using available data. The conditional analysis showed that only rs671 had a significant independent effect on BMI (Supplementary Material, Table S6).

Figure 1.

Regional plots for the four novel loci identified in this study. SNPs are plotted by their position on the chromosome against their association (−log10 P-value) with BMI using Stage I (GWAS meta-analysis) data. The name and P-value for the top SNP shown on the plots is based on all combined data with full genomic control adjustment (Table 1). Estimated recombination rates (from HapMap) are plotted in cyan to reflect the local LD structure. The SNPs surrounding the top SNP (rs671 was used for the ALDH2/MYL2 locus) are color-coded (see inset) to reflect their LD with the top SNP (using pair-wise r2 values from HapMap CHB + JPT data). Genes and positions of exons, as well as directions of transcription, are shown below the plots (using data from the UCSC Genome Browser, genome.ucsc.edu). Plots were generated using LocusZoom.

Figure 1.

Regional plots for the four novel loci identified in this study. SNPs are plotted by their position on the chromosome against their association (−log10 P-value) with BMI using Stage I (GWAS meta-analysis) data. The name and P-value for the top SNP shown on the plots is based on all combined data with full genomic control adjustment (Table 1). Estimated recombination rates (from HapMap) are plotted in cyan to reflect the local LD structure. The SNPs surrounding the top SNP (rs671 was used for the ALDH2/MYL2 locus) are color-coded (see inset) to reflect their LD with the top SNP (using pair-wise r2 values from HapMap CHB + JPT data). Genes and positions of exons, as well as directions of transcription, are shown below the plots (using data from the UCSC Genome Browser, genome.ucsc.edu). Plots were generated using LocusZoom.

To evaluate the possible modifying effect of alcohol consumption on the association between ALDH2 and BMI, we analyzed the association of BMI with rs671 by gender and alcohol consumption status (drinkers versus. non-drinkers) using data from the two studies (SGWAS for Chinese and KCPS-II for Koreans) for which we had direct access to individual data. We found that, among both men and women, the association either was significantly stronger (KCPS-II, P for interaction test = 0.0178) or was only significant (SGWAS) among non-drinkers (Supplementary Material, Table S7).

The ALDH2/SH2B3 locus at 12q24 has been reported to be a target of recent selection in European- and East Asian-ancestry populations (16), with reduction of haplotype diversity. Using the same six representative SNPs (rs4646777, rs671, rs3742000, rs12422941, rs10850014 and rs2301757) reported by Kato et al. (16), we derived the same four common haplotypes (H1, H4, H5, H6) in the two Chinese (SGWAS) and Korean (KCPS-II) data sets mentioned above. The haplotype class specific to East Asians (H5) had the strongest association with BMI in our populations (data not shown).

As shown in Table 2, of the 51 BMI-associated loci that were identified among European-ancestry individuals, the index SNPs at eight loci (rs2890652, rs13078807, rs7638110, rs13107325, rs11847697, rs12444979, rs17024258 and rs10508503) were monomorphic in Asians (Supplementary Material, Table S8). Of the remaining 43 loci, Stage I data revealed that all but one (rs5996074 at SREBF2) had the same direction of association as reported previously (P = 1.0 × 10−11 by the binomial test), eight known loci (near the FTO, BDNF, SEC16B, MC4R, TMEM18, GIPR/QPCTL, ADCY3/RBJ and GNPDA2 genes) were associated with BMI at the genome-wide significance level (P < 5 × 10−8), and another 14 known loci (near the ADCY9, MAP2K5, TFAP2B, TMEM160, OLFM4, FLJ35779, FAIM2, MTCH2, RPL27A, SFRS10/ETV5, NUDT3, HOXB5, ZNF608 and FANCL genes) were associated with BMI at a Bonferroni-corrected significance level (P < 0.05/51 known loci = 1.0 × 10−3). The variation explained by each SNP in these known BMI loci ranged from 0.02–0.15%. The variation explained by all 22 of these re-confirmed BMI-associated loci combined was 1.14%. We compared BMI–SNP associations in East Asian- and European-ancestry populations using data from this study and the GIANT consortium (Supplementary Material, Table S5, S8) and found correlations of effect sizes of r = 0.80 (P = 6.49 × 10−6) for all genome-wide significant loci and r = 0.62 (P = 8.07 × 10−7) for all newly and previously identified loci combined between the two populations.

Table 2.

Associations of SNPs in previously identified loci with BMI in East Asian-ancestry populations

Nearby gene Chr SNP Allelesa EAFb Number of samples β (SE)c Pd Explained variance References 
Eight BMI loci identified in populations of European ancestry were significant at P < 5.0 × 10−8 in East Asian populations 
FTO 16 rs1558902 A/T 0.15 86 668 0.0756 (0.0070) 6.63E−27 0.15% 15,7,8 
BDNF 11 rs11030104 A/G 0.55 86 637 0.0478 (0.0052) 2.36E−20 0.11% 4,8 
SEC16B rs574367 T/G 0.21 86 493 0.0580 (0.0064) 1.93E−19 0.11% 4,8 
MC4R 18 rs591166 A/T 0.24 80 605 0.0464 (0.0062) 7.24E−14 0.08% 35,7,8 
TMEM18 rs12463617 C/A 0.91 84 166 0.0634 (0.0090) 2.08E−12 0.07% 4,5,7,8 
GIPR/QPCTL 19 rs11671664 G/A 0.49 70 606 0.0406 (0.0058) 3.47E−12 0.08% 8,9 
ADCY3/RBJ rs6545814 G/A 0.45 86 669 0.0331 (0.0052) 1.30E−10 0.05% 8,9 
GNPDA2 rs16858082 T/C 0.35 84 150 0.0324 (0.0055) 3.79E−09 0.05% 5,8 
Fourteen BMI loci identified in populations of European ancestry were significant at P < 1.0 × 10−3 in East Asian populations 
ADCY9 16 rs2531995 T/C 0.33 75 987 0.0315 (0.0058) 7.29E−08 0.04% 14 
MAP2K5 15 rs4776970 A/T 0.22 84 217 0.0317 (0.0062) 3.49E−07 0.03% 8,9 
TFAP2B rs9473924 T/G 0.29 76 551 0.0308 (0.0061) 3.77E−07 0.04% 8 
TMEM160 19 rs3810291 A/G 0.24 79 328 0.0333 (0.0068) 8.98E−07 0.04% 8 
OLFM4 13 rs9568867 A/G 0.23 75 149 0.0310 (0.0067) 3.98E−06 0.03% 11 
FLJ35779 rs888789 A/G 0.46 83 977 0.0240 (0.0052) 4.42E−06 0.03% 8 
FAIM2 12 rs897057 C/T 0.79 75 542 0.0287 (0.0068) 2.24E−05 0.03% 8 
MTCH2 11 rs11604680 G/A 0.30 86 354 0.0235 (0.0056) 2.95E−05 0.02% 5,8 
RPL27A 11 rs10160804 A/C 0.47 86 569 0.0212 (0.0051) 3.50E−05 0.02% 8 
SFRS10/ETV5 rs10513801 T/G 0.97 84 121 0.0616 (0.0153) 5.43E−05 0.02% 4,8 
NUDT3 rs4713766 A/C 0.12 61 708 0.0420 (0.0104) 5.49E−05 0.04% 8 
HOXB5 17 rs9299 T/C 0.56 72 384 0.0227 (0.0057) 7.27E−05 0.03% 11 
ZNF608 rs7701094 C/G 0.48 55 908 0.0292 (0.0080) 3.78E−04 0.04% 8 
FANCL rs1861411 A/G 0.41 86 623 0.0183 (0.0053) 5.14E−04 0.02% 12 
Four BMI loci identified in populations of Asian ancestry in the current Stage I meta-analysis 
CDKAL1 rs9356744 T/C 0.57 86 052 0.0374 (0.0052) 5.40E−13 0.07% 9 
PCSK1 rs261967 C/A 0.41 86 488 0.0376 (0.0052) 7.96E−13 0.07% 9 
KLF9 rs11142387 C/A 0.41 70 553 0.0324 (0.0058) 2.79E−08 0.05% 10 
GP2 16 rs12597579 C/T 0.78 86 314 0.0316 (0.0063) 6.13E−07 0.03% 9 
Nearby gene Chr SNP Allelesa EAFb Number of samples β (SE)c Pd Explained variance References 
Eight BMI loci identified in populations of European ancestry were significant at P < 5.0 × 10−8 in East Asian populations 
FTO 16 rs1558902 A/T 0.15 86 668 0.0756 (0.0070) 6.63E−27 0.15% 15,7,8 
BDNF 11 rs11030104 A/G 0.55 86 637 0.0478 (0.0052) 2.36E−20 0.11% 4,8 
SEC16B rs574367 T/G 0.21 86 493 0.0580 (0.0064) 1.93E−19 0.11% 4,8 
MC4R 18 rs591166 A/T 0.24 80 605 0.0464 (0.0062) 7.24E−14 0.08% 35,7,8 
TMEM18 rs12463617 C/A 0.91 84 166 0.0634 (0.0090) 2.08E−12 0.07% 4,5,7,8 
GIPR/QPCTL 19 rs11671664 G/A 0.49 70 606 0.0406 (0.0058) 3.47E−12 0.08% 8,9 
ADCY3/RBJ rs6545814 G/A 0.45 86 669 0.0331 (0.0052) 1.30E−10 0.05% 8,9 
GNPDA2 rs16858082 T/C 0.35 84 150 0.0324 (0.0055) 3.79E−09 0.05% 5,8 
Fourteen BMI loci identified in populations of European ancestry were significant at P < 1.0 × 10−3 in East Asian populations 
ADCY9 16 rs2531995 T/C 0.33 75 987 0.0315 (0.0058) 7.29E−08 0.04% 14 
MAP2K5 15 rs4776970 A/T 0.22 84 217 0.0317 (0.0062) 3.49E−07 0.03% 8,9 
TFAP2B rs9473924 T/G 0.29 76 551 0.0308 (0.0061) 3.77E−07 0.04% 8 
TMEM160 19 rs3810291 A/G 0.24 79 328 0.0333 (0.0068) 8.98E−07 0.04% 8 
OLFM4 13 rs9568867 A/G 0.23 75 149 0.0310 (0.0067) 3.98E−06 0.03% 11 
FLJ35779 rs888789 A/G 0.46 83 977 0.0240 (0.0052) 4.42E−06 0.03% 8 
FAIM2 12 rs897057 C/T 0.79 75 542 0.0287 (0.0068) 2.24E−05 0.03% 8 
MTCH2 11 rs11604680 G/A 0.30 86 354 0.0235 (0.0056) 2.95E−05 0.02% 5,8 
RPL27A 11 rs10160804 A/C 0.47 86 569 0.0212 (0.0051) 3.50E−05 0.02% 8 
SFRS10/ETV5 rs10513801 T/G 0.97 84 121 0.0616 (0.0153) 5.43E−05 0.02% 4,8 
NUDT3 rs4713766 A/C 0.12 61 708 0.0420 (0.0104) 5.49E−05 0.04% 8 
HOXB5 17 rs9299 T/C 0.56 72 384 0.0227 (0.0057) 7.27E−05 0.03% 11 
ZNF608 rs7701094 C/G 0.48 55 908 0.0292 (0.0080) 3.78E−04 0.04% 8 
FANCL rs1861411 A/G 0.41 86 623 0.0183 (0.0053) 5.14E−04 0.02% 12 
Four BMI loci identified in populations of Asian ancestry in the current Stage I meta-analysis 
CDKAL1 rs9356744 T/C 0.57 86 052 0.0374 (0.0052) 5.40E−13 0.07% 9 
PCSK1 rs261967 C/A 0.41 86 488 0.0376 (0.0052) 7.96E−13 0.07% 9 
KLF9 rs11142387 C/A 0.41 70 553 0.0324 (0.0058) 2.79E−08 0.05% 10 
GP2 16 rs12597579 C/T 0.78 86 314 0.0316 (0.0063) 6.13E−07 0.03% 9 

aShown as effect allele/other allele.

bEffect allele frequency, estimated from Stages I and II studies for Asians.

cPer allele effects of SNPs on BMI are presented in standard deviations, which were derived from the meta-analysis.

dDerived from the meta-analysis and adjusted for both study-specific inflation factors (for Stages I and II) and for the estimated inflation factor for the Stage I meta-analysis statistic.

To compare the genetic architecture of regions associated with BMI between Asians and Europeans, we investigated the linkage disequilibrium (LD; by r2) of SNPs in the 200 kb flanking all previously (Supplementary Material, Table S8) and newly (Table 1) identified BMI loci in both populations. We calculated the pairwise distance and LD (r2) for each locus in each population based on HapMap3 SNP data through the public SNP Annotation and Proxy Search (SNAP) tool. The average LD decay over distance for the two populations showed similar patterns, suggesting that the genetic structure of those regions is similar (Supplementary Material, Fig. S2).

The reported effect sizes for all BMI-related SNPs in studies of European-ancestry populations are usually >3% of the standard deviation of BMI (4). Given the size of our study (N = 86 757 for Stage I), we had adequate statistical power (>80% at a significance level of P < 1.0 × 10−3) to detect a SNP with such an effect size and a MAF of >0.12. Previously reported loci that were not replicated in our study at P < 1.0 × 10−3 had either a very small effect size or a low MAF (Supplementary Material, Table S8).

Of the four BMI-associated loci we identified in our previous studies conducted among East Asians (9,10), Stage I data showed that 3 loci (in the PCSK1, CDKAL1 and KLF9 genes) remained genome-wide significant (P < 5.0 × 10−8), while the GP2 locus did not reach the genome-wide significance level (P = 6.13 × 10−7) (Table 2, Supplementary Material, Table S8). The variation explained by all four of these loci combined was 0.22%. Altogether, the overall variation explained by the 30 re-confirmed or newly identified BMI-associated loci (22 loci originally identified in Europeans, 4 loci originally identified in East Asians and 4 newly identified loci) was 1.52%, which is an improvement over the previously reported value of 1.18% in East Asians (9). Assuming that the 21 BMI loci identified in European-ancestry populations that we did not confirm in this study could be confirmed with a larger sample size, the variation explained by all known BMI loci would be 1.65%. We anticipate that the variation explained by genetics will increase when rare variants are considered.

Additional analyses examined effect sizes for differences across sex, population, individual studies and obesity status. Analyses stratified by sex (Table 3) showed that associations with BMI among men were significantly stronger than associations among women for rs2237892 in KCNQ1 (effect size: 0.0411 versus 0.0204, P for homogeneity = 1.07 × 10−2), rs671 in ALDH2 (effect size: 0.0560 versus 0.0234, P for homogeneity = 3.11 × 10−3) and rs12229654 in MYL2 (effect size: 0.0543 versus 0.0190, P for homogeneity = 1.76 × 10−3). In addition, we also observed a stronger association among men than among women in two of our previously reported loci at CDKAL1 (P for homogeneity = 5.74 × 10−3) and PCSK1 (P for homogeneity = 5.95 × 10−3) (Supplementary Material, Table S8). Analyses stratified by population (Supplementary Material, Table S9) showed that associations with BMI for all four new loci were similar (P for homogeneity ≥ 0.15) across Chinese, Japanese and Korean populations, although none were statistically significant among Malay/Filipino populations. No significant heterogeneity across individual studies was found for these four new loci (data not shown). Meta-analyses of obesity as a dichotomous outcome (BMI ≥ 27.5 kg/m2) (17) also showed similar associations with odds ratios per allele ranging from 1.03 to 1.09, although the statistical power for this analysis was lower (Supplementary Material, Table S10).

Table 3.

Newly identified loci associated with BMI variation in East Asian-ancestry populations, by gender

Nearby gene Chr SNP Allelesa Among men
 
Among women
 
Test for homogeneity 
    Number β (SE)b Pc Number β (SE)b Pc P 
KCNQ1 11 rs2237892 T/C 59 365 0.0411 (0.0059) 4.54E−12 72 300 0.0204 (0.0055) 2.18E−04 1.07E−02 
ALDH2 12 rs671 G/A 42 896 0.0560 (0.0080) 1.97E−12 53 421 0.0234 (0.0077) 2.32E−03 3.11E−03 
MYL2 12 rs12229654 T/G 48 395 0.0543 (0.0083) 5.45E−11 60 141 0.0190 (0.0077) 1.38E−02 1.76E−03 
ITIH4 rs2535633 G/C 48 927 0.0289 (0.0065) 8.29E−06 61 184 0.0266 (0.0059) 6.13E−06 7.96E−01 
NT5C2 10 rs11191580 C/T 42 636 0.0252 (0.0079) 1.47E−03 53 382 0.0332 (0.0070) 2.03E−06 4.49E−01 
Nearby gene Chr SNP Allelesa Among men
 
Among women
 
Test for homogeneity 
    Number β (SE)b Pc Number β (SE)b Pc P 
KCNQ1 11 rs2237892 T/C 59 365 0.0411 (0.0059) 4.54E−12 72 300 0.0204 (0.0055) 2.18E−04 1.07E−02 
ALDH2 12 rs671 G/A 42 896 0.0560 (0.0080) 1.97E−12 53 421 0.0234 (0.0077) 2.32E−03 3.11E−03 
MYL2 12 rs12229654 T/G 48 395 0.0543 (0.0083) 5.45E−11 60 141 0.0190 (0.0077) 1.38E−02 1.76E−03 
ITIH4 rs2535633 G/C 48 927 0.0289 (0.0065) 8.29E−06 61 184 0.0266 (0.0059) 6.13E−06 7.96E−01 
NT5C2 10 rs11191580 C/T 42 636 0.0252 (0.0079) 1.47E−03 53 382 0.0332 (0.0070) 2.03E−06 4.49E−01 

aShown as effect allele/other allele.

bPer-allele effects of SNPs on BMI are presented in standard deviations, which were derived from the meta-analysis.

cDerived from the meta-analysis and adjusted for both study-specific inflation factors (for Stages I and II) and the estimated inflation factor for the Stage I meta-analysis statistic.

In an effort to search for potential functional variants, we systemically examined expression quantitative trait loci (eQTL) in the 1 Mb regions flanking the four newly identified loci. A total of 178 eQTLs (Supplementary Material, Table S11) were identified in public databases and the previous literature. We next investigated whether these eQTL SNPs were located in certain functional elements using the online tool HaploReg (18). We found that of the 178 eQTL SNPs, 69.7% were located in enhancer regions. This percentage is significantly higher (P = 2.2 × 10−16) than the percentage of enhancer regions in the human genome (19.8%). In particular, the four newly identified loci are all located in motif binding sites and are associated with enhancer regions (Supplementary Material, Table S12).

To further explore over-represented biological pathways among the genes located near the newly and previously identified BMI loci listed in Table 1 and Supplementary Material, Table S8, we examined their functional enrichment in biological pathway analyses using the ingenuity pathway analysis (IPA) tool in Ingenuity (version 17199142). We found that two relevant BMI pathways, CDK5 signaling (P = 1.94 × 10−4) and corticotropin-releasing hormone signaling (P = 3.74 × 10−4), were significantly enriched.

DISCUSSION

Of the four newly identified BMI-associated loci in this study, SNP rs2237892 is located in an intron of the KCNQ1 gene, which encodes a voltage-gated potassium channel. This locus is involved in long QT syndrome in Europeans and African Americans (19,20) and is associated with type 2 diabetes (T2D) in both Asian and European populations (2123). The T2D risk-associated C allele of rs2237892 has been related to lower fasting insulin levels (24) and a reduction in insulin secretion (25). The current study found that this risk allele is also associated with lower BMI. Adjusting for BMI in logistic regression models has been shown to strengthen rather than attenuate the association of rs2237892 with T2D (26). Given the strong link between T2D and obesity, we carried out additional analyses after excluding participants with T2D and found that the association of rs2237892 with BMI remained (P = 3.72 × 10−8). While the relationships of T2D with insulin secretion and insulin resistance are clear, the cause-and-effect relationships between hyperinsulinemia, insulin resistance, obesity and T2D remain unresolved. One study has suggested that suppression of insulin secretion was associated with loss of body weight and fat mass (27).

The locus represented by rs671 contains the ALDH2 gene, which is involved in dehydrogenation of acetaldehyde and is associated with alcohol consumption behavior and alcohol-flushing responses in Asians (22,28,29). GWAS have reported that the BMI-increasing allele of this SNP is associated with diverse traits, including alcohol consumption behavior (22), increased intracranial aneurysm (30), triglycerides (31), gamma glutamyl transferase levels (32), elevated blood pressure (16), lower risk of coronary heart disease (33), decreased alcohol-flushing responses and esophageal cancer (34). rs671 results in a glutamine to lysine missense change at position 504 in the ALDH2 protein (accession ID NP_000681.2), known as the ALDH*2 allele, and is predicted by both PolyPhen-2 (35) and SIFT (36) to be functionally important. A recent Mendelian randomization study suggested that ALDH2 may influence the risk of hypertension by affecting alcohol consumption behavior, with ALDH*1 allele carriers having higher blood pressure due to higher alcohol consumption (37). However, our study (Supplementary Material, Table S7) suggested an antagonistic effect of alcohol consumption on the ALDH2–BMI association. The ALDH*1 BMI-increasing effect was mainly observed among non-drinkers.

While rs671 appears to be the most likely candidate in the 12q24 region, it is also in strong LD with the A allele of rs3782886 (r2 = 0.95), which reached the genome-wide significance level in our Stage I data (P = 1.24 × 10−8) and is associated with decreased levels of alanine aminotransferase (32). Although its association with BMI was no longer significant after adjustment for rs671 in our study, another SNP in the 12q24 region, rs12229654 near the MYL2 gene, has been associated with HDL cholesterol (38), levels of gamma glutamyl transpeptidase (38) and alcohol consumption (39) in Asian-ancestry populations. SNP rs12229654 is in LD (r2 = 0.67) with 3 SNPs (rs11065756, rs3782888 and rs12231049) that are predicted to be among the strongest eQTLs in the region in HapMap lymphoblastoid cell lines for the MYL2 gene (40) (P < 0.05, Supplementary Material, Table S11). MYL2 encodes the myosin light chain and is involved in heart morphogenesis, and downregulation of this gene has been posited to play a role in coronary artery disease (41). In a Korean population, new loci in MYL2 were recently shown to be associated with plasma glucose levels (42) and HDL levels (38). A SNP in the 12q24 region that is in LD (r2 = 0.58) with rs671, rs2074356, has been previously associated with waist-to-hip ratio (43).

The third new locus, rs2535633, is in an intron of the ITIH4 gene, which has been reported to be involved in the stabilization of the extracellular matrix and shows wide expression in the blood and liver (44). Obesity in rats has been positively correlated with rat blood levels of the ITIH4 protein, which has led to the suggestion that this protein may act as a biomarker for obesity (45). Fujita et al. (46) reported an association of the ITIH4 gene with total cholesterol levels in individuals of Japanese ancestry. SNP rs2535633 is in LD with two non-synonymous SNPs in the ITIH4 gene, rs13072536 and rs4687657 (r2 = 0.83 and 0.71, respectively), that reached the genome-wide significance level in Stage I (P = 2.05 × 10−8 for rs13072536 and P = 2.63 × 10−8 for rs4687657). Whereas rs13072536 is predicted by PolyPhen-2 (35) to be ‘probably damaging’, rs4687657 is predicted to be ‘damaging’ by SIFT (36). SNP rs2535633 is also an eQTL in HapMap lymphoblastoid cell lines for the ITIH4 (P = 5.5 × 10−7), FLJ12442 (P = 1.7 × 10−6) and TMEM110 (P = 2.2 × 10−19) (47,48) genes and is in strong LD with other SNPs also predicted to act as eQTLs in lymphoblastoid cell lines and monocytes for ITIH4, ITIH3, NT5DC2, WRD51A and FLJ12442 (40,4749). This, in combination with biomarker studies in rats suggest that ITIH4 levels (45), which may be higher in those with the risk allele, may help identify individuals at risk for obesity. In addition, rs11918800 (r2 = 1.0 with rs2535633) is located in a predicted transcription factor binding site, and rs6445538 (r2 = 0.73 with rs2535633) is in a predicted hsa-miR-1301 miRNA binding site (50). The precise mechanisms by which one or more of these SNPs act on gene function and BMI remain to be determined.

Finally, the index SNP for the fourth new locus, rs11191580, resides in an intron of the NT5C2 gene and has been associated with a number of psychiatric disorders, including autism and schizophrenia (5153). Another SNP, rs11191548, which is in complete LD with rs11191580 (r2 = 1), has been associated with measures of blood pressure in both European- and Asian-ancestry populations in four previous GWAS (16,5456). Genetic variations in this gene were recently found to be associated with reduced subcutaneous and visceral fat mass in Japanese women (57). Further, rs11191580 is in strong LD with a number of SNPs that are predicted eQTLs for the USMG5 gene according to two different datasets [P = 4.5 × 10−7 by Veyrieras et al. (40), P = 9.7 × 10−55 by Zeller et al. (47)]. The USMG5 gene has been identified as coding a diabetes-associated protein in insulin-sensitive tissue (58). A recent study (59) reported a locus (rs12413409) that was associated with coronary artery disease. This SNP is in strong LD with rs11191580 (r2 = 1 in Europeans, r2 = 0.895 in Asians) and was associated with BMI (P = 6.67 × 10−7) in our Stage I data.

We observed similarities in the genetic architecture of BMI loci between Asian- and European-ancestry populations, despite notable differences in allele frequencies for some BMI loci, such as loci that were monomorphic. However, BMI distribution in Asians is very different from that in Europeans, supporting the notion that non-genetic factors, such as diet and physical activity, play a more important role in obesity than genetic factors. In fact, only a small percentage of BMI variation can be explained by genetic loci (1.52% in Asians). Clearly, further research is needed to investigate the interaction between genetic and lifestyle factors on the worldwide obesity epidemic.

The eQTL analysis suggested evidence of a potential functional role for the newly identified loci. Pathway analysis found two BMI-related pathways. One is cyclin-dependent kinase (CDK5) signaling, which can result in phosphorylation of the nuclear receptor PPARγ, which is encoded by the PPARG gene, a ‘master’ gene for fat cell biology and differentiation (60,61). Another top pathway was corticotropin-releasing hormone signaling (P = 3.74 × 10−4), which has been associated with depression and type 2 diabetes (62). A more thorough investigation and experimental verification are warranted to definitively establish the causal connections.

It is worth noting that four of the newly identified BMI-associated loci, KCNQ1, ALDH2, ITIH4 and NT5C2, showed substantial pleiotropic effects, as mentioned above, on multiple obesity-related chronic-disease traits, such as T2D, blood pressure, coronary heart disease and schizophrenia. Of note, the BMI-decreasing alleles are associated with increased risk of T2D (KCNQ1), elevated blood pressure (NT5C2) and schizophrenia (ITIH4 and NT5C2). However, the BMI-decreasing allele of rs671 in ALDH2 is associated with decreased blood pressure and increased risk of coronary heart disease. Further studies are warranted to elaborate on the causal relationship between these genes, chronic-disease traits and obesity.

In conclusion, our study confirmed 22 previously reported BMI-associated loci in studies of European-ancestry populations and identified four novel loci near the KCNQ1, ALDH2/MYL2, ITIH4 and NT5C2 genes that are associated with BMI at the genome-wide significance level. The SNPs in the KCNQ1 and ALDH2/MYL2 genes showed stronger effects among men compared with women. SNPs rs671 and rs12229654 in ALDH2/MYL2 are monomorphic in European-ancestry populations. Our study demonstrates the value of conducting genetic studies in different ethnic populations and expands our knowledge of the genetic basis for obesity.

MATERIALS AND METHODS

Study design

This study had two stages. Stage I was a meta-analysis of study-specific results on the association between SNPs and BMI from the 21 GWAS that participated in the consortium and included a total of 86 757 individuals of Asian ancestry. Promising SNPs selected from the Stage I meta-analysis were further examined by de novo or in silico replication analyses (Stage II). Supplementary Material, Tables S1–3, Figure S1 and the Supplementary Information summarize the basic information for all participating studies.

Stage I samples and genotyping

The sample sizes of the 21 GWAS in Stage I varied from 821 to 33 530, with a total of 86 757 individuals. Nine studies used Affymetrix arrays, and 12 studies used the Illumina platform (detailed information is provided in the Supplementary Information). To allow for combination of the data derived from different genotyping platforms and to improve coverage of the genome, genotype imputation was performed by each participating study using either MACH or IMPUTE with HapMap CHD + JPT data (release #22, build 36) as the imputation reference panel (Supplementary Material, Table S2).

Stage I statistical analysis

A uniform statistical analysis protocol was followed by each participating study. BMI was calculated by dividing weight in kilograms by the square of height in meters. To improve the normality of the BMI distribution and alleviate the impact of outliers, rank-based inverse normal transformation (INT) was applied to BMI data separately for each gender by each study. INT involves ranking all BMI values, transforming these ranks into quantiles and, finally, converting the resulting quantiles into normal deviates. Associations between SNPs and the inverse normal-transformed BMI were analyzed with a linear regression model; associations between SNPs and obesity were analyzed as a dichotomous outcome, in which obesity was defined as BMI ≥ 27.5 (17), by using a logistic regression model, assuming an underlying additive genetic model and adjusting for age (continuous), age-squared and gender (if applicable). Stratified analyses by gender and disease status (with or without cancer and T2D) were also performed by each study.

Next, we carried out meta-analyses using a weighted average method with inverse-variance weights. The meta-analyses were carried out on all data combined and also stratified by gender and disease status using the freely available METAL software. The presence of heterogeneity across studies and between genders was tested with Cochran's Q statistics (63).

To correct each study for residual population stratification or cryptic relatedness, the meta-analyses were performed with genomic control correction (64) by adjusting for the study-specific inflation factor (λ), which ranged from 1.000 to 1.123 in Stage I (Supplementary Material, Table S2). After study-specific genomic control adjustment, the estimated inflation factor for the Stage I meta-analysis statistic was 1.128, which was further adjusted for when calculating the Stage I results.

Stage II replication analysis

Eight SNPs that were not near any previously reported BMI-associated loci and that had P < 7.59 × 10−6 in the Stage I data were taken forward into the Stage II replication analysis. The Stage II studies included a total of 47 791 individuals and consisted of de novo genotyping data from three study sites and in silico replication data from the Tai Chi study, which had been previously genotyped with Illumina's iSelect 200k Cardio-MetaboChip (Supplementary Material, Table S3). Due to the differing availability of replication data, for each SNP the sample size for the Stage II analysis varied from 7488 for rs4854307 to 47 352 for rs2237892.

Each study individually conducted a similar analysis of the association between BMI and the selected SNPs, using the same protocol used in Stage I. The Stage II data were combined using the same meta-analysis methods as in Stage I. Finally, we used meta-analysis to combine all data from both Stages I and II.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

URLS

METAL program, http://www.sph.umich.edu/csg/abecasis/Metal/; Cardio-MetaboChip, http://www.sph.umich.edu/csg/kang/MetaboChip/; HaploReg, http://www.broadinstitute.org/mammals/haploreg/haploreg.php/; Ingenuity, http://www.ingenuity.com/. SNAP, http://www.broadinstitute.org/mpg/snap/ldsearchpw.php

FUNDING

This work was supported by the sources listed below. The funding information provided below pertains to the participating studies that contributed summary statistics to this meta-analysis. The funders of the original studies had no role in study design, data collection and analysis, decision to publish or preparation of this manuscript. The SGWAS was supported in part by US National Institutes of Health grants R37CA070867 (to W.Z.), R01CA082729 (to X.-O. S.), R01CA124558 (to W.Z.), R01CA148667 (to W.Z.) and R01CA122364 (G.Y.), as well as Ingram Professorship and Research Reward funds from the Vanderbilt University School of Medicine. Participating studies (grant support) in the Shanghai Genome-Wide Association Studies (SGWAS) are as follows: Shanghai Women's Health Study (R37CA070867 to W.Z.), Shanghai Men′s Health Study (R01CA082729 to X.-O.S.), Shanghai Breast Cancer Study (R01CA064277 to X.-O.S.) and Shanghai Endometrial Cancer Study (R01CA092585 to X.-O.S.). We thank Regina Courtney for DNA preparation and Jing He for data processing and analyses. The JMGP was supported by Grants for Scientific Research (Priority Areas “Medical Genome Science (Millennium Genome Project)” and “Applied Genomics”, Leading Project for Personalized Medicine, and Scientific Research 20390185, 21390099, 19659163, 16790336, 12204008, 15790293, 16590433, 17790381, 17790381, 18390192, 18590265, 18590587, 18590811, 19590929, 19650188, 19790423, 17390186, 20390184 and 21390223) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan; a Grants-in-Aid [H15-Longevity-005, H17-longevity-003, H16-kenko-001, H18-longevity (kokusai), H11-longevity-020, H17-Kenkou-007, H17-pharmaco-common-003, H18-Junkankitou (Seishuu)-Ippan-012 and H20-Junkankitou (Seishuu)-Ippan-009, 013] from the Ministry of Health, Labor and Welfare, Health and Labor Sciences Research Grants, Japan; a Science and Technology Incubation Program in Advanced Regions, Japan Science and Technology Agency; a Grants-in-Aid from the Japan Society for the Promotion of Science (JSPS) fellows (16.54041, 18.54042, 19.7152, 20.7198, 20.7477 and 20.54043), Tokyo, Japan; Health Science Research Grants and Medical Technology Evaluation Research Grants from the Ministry of Health, Labor and Welfare, Japan; the Japan Atherosclerosis Prevention Fund; the Uehara Memorial Foundation; the Takeda Medical Research Foundation; and the Japan Research Foundation for Clinical Pharmacology. The KARE project was supported by a grant from the Korea Center for Disease Control and Prevention (4845-301, 4851-302, 4851-307), and intramural grant from the Korea National Institute of Health (2012-N73002-00). The SP2 and SiMES were supported by the Singapore Ministry of Health′s National Medical Research Council under its Individual Research Grant funding scheme, the Singapore National Research Foundation under its Clinician Scientist Award and Singapore Translational Research Investigator Award funding schemes, which are administered by the Singapore Ministry of Health′s National Medical Research Council and the Singapore Biomedical Research Council (BMRC) individual research grant funding scheme. The NHAPC study is supported by research grants including the National High Technology Research and Development Program (2009AA022704), Knowledge Innovation Program (KSCX2-EW-R-10), the National Natural Science Foundation of China (30930081, 81021002, 81170734) and the National Key Basic Research Program of China (2012CB524900). GenSalt is supported by research grants (U01HL072507, R01HL087263 and R01HL090682) from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD. The CLHNS was supported by National Institutes of Health grants DK078150, TW05596, HL085144 and TW008288 and pilot funds from RR20649, ES10126 and DK56350. We thank the Office of Population Studies Foundation research and data collection teams for the Cebu Longitudinal Health and Nutrition Survey. The CRC was supported by grants from the National Heart, Lung, and Blood Institute (HL071981), the National Institute of Diabetes and Digestive and Kidney Diseases (DK091718 and DK078616), the Boston Obesity Nutrition Research Center (DK46200) and United States—Israel Binational Science Foundation Grant 2011036. The MESA and MESA SHARe project are conducted and supported by contracts N01-HC-95159 through N01-HC-95169 and RR-024156 from the National Heart, Lung, and Blood Institute (NHLBI). Funding for MESA SHARe genotyping was provided by NHLBI Contract N02-HL-6-4278. The authors thank the participants of the MESA study, the Coordinating Center, MESA investigators and study staff for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. The TAICHI Metabochip study was supported by NHLBI grant HL087647. Financial support for HALST was through grants from the National Health Research Institutes (PH-100-SP-01). The SAPPHIRe was supported by grants from the National Health Research Institutes (BS-094-PP-01 & PH-100-PP-03). The TCAGEN was partially supported by grants NTUH.98-N1266, NTUH100-N1775, NTUH101-N2010, NTUH101-N, VN101-04 and NTUH 101-S1784 from National Taiwan University Hospital, NSC 96-2314-B-002-152 and NSC 101-2325-002-078. The TACT was supported by grants from the National Science Council of Taiwan (NSC96-2314-B-002-151, NSC98-2314-B-002-122-MY2 and NSC 100-2314-B-002-115). The Taiwan Dragon and TACD were supported by grants from the National Science Council (NSC 98-2314-B-075A-002-MY3) and Taichung Veterans General Hospital, Taichung, Taiwan (TCVGH-1013001C; TCVGH-1013002D). The Taiwan Genome Wide Association Study was supported by the Academia Sinica Genomic Medicine Multicenter Study (Academia Sinica 40-05-GMM) and Search and build the biosignatures for type 2 diabetes complications in the Han Chinese population (Academia Sinica 23-2 h), Academia Sinica, Taiwan, and the National Center for Genome Medicine at Academia Sinica (NCGM, NSC-101-2319-B-001-001) of the National Core Facility Program for Biotechnology (NCFPB) and the Translational Resource Center for Genomic Medicine (TRC, NSC-101-2325-B-001-035) of the National Research Program for Biopharmaceuticals (NRPB), National Science Council, Taiwan.

ACKNOWLEDGEMENTS

The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies. The authors wish to thank the study participants and research staff of each contributing study for their contributions and commitment, which made this project possible, and Bethanie Rammer for editing and preparing the manuscript.

Conflict of Interest statement. None declared.

REFERENCES

1
Frayling
T.M.
Timpson
N.J.
Weedon
M.N.
Zeggini
E.
Freathy
R.M.
Lindgren
C.M.
Perry
J.R.
Elliott
K.S.
Lango
H.
Rayner
N.W.
et al.  
A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity
Science
 
2007
316
889
894
2
Scuteri
A.
Sanna
S.
Chen
W.M.
Uda
M.
Albai
G.
Strait
J.
Najjar
S.
Nagaraja
R.
Orru
M.
Usala
G.
et al.  
Genome-wide association scan shows genetic variants in the FTO gene are associated with obesity-related traits
PLoS Genet.
 
2007
3
e115
3
Loos
R.J.
Lindgren
C.M.
Li
S.
Wheeler
E.
Zhao
J.H.
Prokopenko
I.
Inouye
M.
Freathy
R.M.
Attwood
A.P.
Beckmann
J.S.
et al.  
Common variants near MC4R are associated with fat mass, weight and risk of obesity
Nat. Genet.
 
2008
40
768
775
4
Thorleifsson
G.
Walters
G.B.
Gudbjartsson
D.F.
Steinthorsdottir
V.
Sulem
P.
Helgadottir
A.
Styrkarsdottir
U.
Gretarsdottir
S.
Thorlacius
S.
Jonsdottir
I.
et al.  
Genome-wide association yields new sequence variants at seven loci that associate with measures of obesity
Nat. Genet.
 
2009
41
18
24
5
Willer
C.J.
Speliotes
E.K.
Loos
R.J.
Li
S.
Lindgren
C.M.
Heid
I.M.
Berndt
S.I.
Elliott
A.L.
Jackson
A.U.
Lamina
C.
et al.  
Six new loci associated with body mass index highlight a neuronal influence on body weight regulation
Nat. Genet.
 
2009
41
25
34
6
Meyre
D.
Delplanque
J.
Chevre
J.C.
Lecoeur
C.
Lobbens
S.
Gallina
S.
Durand
E.
Vatin
V.
Degraeve
F.
Proenca
C.
et al.  
Genome-wide association study for early-onset and morbid adult obesity identifies three new risk loci in European populations
Nat. Genet.
 
2009
41
157
159
7
Scherag
A.
Dina
C.
Hinney
A.
Vatin
V.
Scherag
S.
Vogel
C.I.
Muller
T.D.
Grallert
H.
Wichmann
H.E.
Balkau
B.
et al.  
Two new Loci for body-weight regulation identified in a joint analysis of genome-wide association studies for early-onset extreme obesity in French and German study groups
PLoS Genet.
 
2010
6
e1000916
8
Speliotes
E.K.
Willer
C.J.
Berndt
S.I.
Monda
K.L.
Thorleifsson
G.
Jackson
A.U.
Allen
H.L.
Lindgren
C.M.
Luan
J.
Magi
R.
et al.  
Association analyses of 249,796 individuals reveal 18 new loci associated with body mass index
Nat. Genet.
 
2010
42
937
948
9
Wen
W.
Cho
Y.S.
Zheng
W.
Dorajoo
R.
Kato
N.
Qi
L.
Chen
C.H.
Delahanty
R.J.
Okada
Y.
Tabara
Y.
et al.  
Meta-analysis identifies common variants associated with body mass index in east Asians
Nat. Genet.
 
2012
44
307
311
10
Okada
Y.
Kubo
M.
Ohmiya
H.
Takahashi
A.
Kumasaka
N.
Hosono
N.
Maeda
S.
Wen
W.
Dorajoo
R.
Go
M.J.
et al.  
Common variants at CDKAL1 and KLF9 are associated with body mass index in east Asian populations
Nat. Genet.
 
2012
44
302
306
11
Bradfield
J.P.
Taal
H.R.
Timpson
N.J.
Scherag
A.
Lecoeur
C.
Warrington
N.M.
Hypponen
E.
Holst
C.
Valcarcel
B.
Thiering
E.
et al.  
A genome-wide association meta-analysis identifies new childhood obesity loci
Nat. Genet.
 
2012
44
526
531
12
Guo
Y.
Lanktree
M.B.
Taylor
K.C.
Hakonsarson
H.
Lange
L.A.
Keating
B.J.
Gene-centric meta-analyses of 108 912 individuals confirm known body mass index loci and reveal three novel signals
Hum. Mol. Genet.
 
2013
22
184
201
13
Melka
M.G.
Bernard
M.
Mahboubi
A.
Abrahamowicz
M.
Paterson
A.D.
Syme
C.
Lourdusamy
A.
Schumann
G.
Leonard
G.T.
Perron
M.
et al.  
Genome-wide scan for loci of adolescent obesity and their relationship with blood pressure
J. Clin. Endocrinol. Metab.
 
2012
97
E145
E150
14
Berndt
S.I.
Gustafsson
S.
Magi
R.
Ganna
A.
Wheeler
E.
Feitosa
M.F.
Justice
A.E.
Monda
K.L.
Croteau-Chonka
D.C.
Day
F.R.
et al.  
Genome-wide meta-analysis identifies 11 new loci for anthropometric traits and provides insights into genetic architecture
Nat. Genet.
 
2013
45
501
512
15
Randall
J.C.
Winkler
T.W.
Kutalik
Z.
Berndt
S.I.
Jackson
A.U.
Monda
K.L.
Kilpelainen
T.O.
Esko
T.
Magi
R.
Li
S.
et al.  
Sex-stratified genome-wide association studies including 270,000 individuals show sexual dimorphism in genetic loci for anthropometric traits
PLoS Genet.
 
2013
9
e1003500
16
Kato
N.
Takeuchi
F.
Tabara
Y.
Kelly
T.N.
Go
M.J.
Sim
X.
Tay
W.T.
Chen
C.H.
Zhang
Y.
Yamamoto
K.
et al.  
Meta-analysis of genome-wide association studies identifies common variants associated with blood pressure variation in east Asians
Nat. Genet.
 
2011
43
531
538
17
WHO expert consultation
Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies
Lancet
 
2004
363
157
163
18
Yeates
T.O.
Determination of the correct reference frame from an atomic coordinate list
Acta Crystallogr. A
 
1990
46
Pt 7
625
626
19
Smith
J.G.
Avery
C.L.
Evans
D.S.
Nalls
M.A.
Meng
Y.A.
Smith
E.N.
Palmer
C.
Tanaka
T.
Mehra
R.
Butler
A.M.
et al.  
Impact of ancestry and common genetic variants on QT interval in African Americans
Circ. Cardiovasc. Genet.
 
2012
5
647
655
20
Pfeufer
A.
Sanna
S.
Arking
D.E.
Muller
M.
Gateva
V.
Fuchsberger
C.
Ehret
G.B.
Orru
M.
Pattaro
C.
Kottgen
A.
et al.  
Common variants at ten loci modulate the QT interval duration in the QTSCD Study
Nat. Genet.
 
2009
41
407
414
21
Voight
B.F.
Scott
L.J.
Steinthorsdottir
V.
Morris
A.P.
Dina
C.
Welch
R.P.
Zeggini
E.
Huth
C.
Aulchenko
Y.S.
Thorleifsson
G.
et al.  
Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis
Nat. Genet.
 
2010
42
579
589
22
Takeuchi
F.
Isono
M.
Nabika
T.
Katsuya
T.
Sugiyama
T.
Yamaguchi
S.
Kobayashi
S.
Ogihara
T.
Yamori
Y.
Fujioka
A.
Confirmation of ALDH2 as a Major locus of drinking behavior and of its variants regulating multiple metabolic phenotypes in a Japanese population
Circ. J.
 
2011
75
911
23
Yasuda
K.
Miyake
K.
Horikawa
Y.
Hara
K.
Osawa
H.
Furuta
H.
Hirota
Y.
Mori
H.
Jonsson
A.
Sato
Y.
Variants in KCNQ1 are associated with susceptibility to type 2 diabetes mellitus
Nat. Genet.
 
2008
40
1092
1097
24
Dai
X.P.
Huang
Q.
Yin
J.Y.
Guo
Y.
Gong
Z.C.
Lei
M.X.
Jiang
T.J.
Zhou
H.H.
Liu
Z.Q.
KCNQ1 gene polymorphisms are associated with the therapeutic efficacy of repaglinide in Chinese Type 2 diabetic patients
Clin. Exp. Pharmacol. Physiol.
 
2012
39
462
468
25
Tan
J.T.
Nurbaya
S.
Gardner
D.
Ye
S.
Tai
E.S.
Ng
D.P.
Genetic variation in KCNQ1 associates with fasting glucose and beta-cell function: a study of 3,734 subjects comprising three ethnicities living in Singapore
Diabetes
 
2009
58
1445
1449
26
Saif-Ali
R.
Ismail
I.S.
Al-Hamodi
Z.
Al-Mekhlafi
H.M.
Siang
L.C.
Alabsi
A.M.
Muniandy
S.
KCNQ1 Haplotypes associate with Type 2 Diabetes in Malaysian Chinese subjects
Int. J. Mol. Sci.
 
2011
12
5705
5718
27
Velasquez-Mieyer
P.A.
Cowan
P.A.
Arheart
K.L.
Buffington
C.K.
Spencer
K.A.
Connelly
B.E.
Cowan
G.W.
Lustig
R.H.
Suppression of insulin secretion is associated with weight loss and altered macronutrient intake and preference in a subset of obese adults
Int. J. Obes. Relat. Metab. Disord.
 
2003
27
219
226
28
Yoshida
A.
Huang
I.Y.
Ikawa
M.
Molecular abnormality of an inactive aldehyde dehydrogenase variant commonly found in Orientals
Proc. Natl. Acad. Sci.
 
1984
81
258
261
29
Wang
Y.
Zhang
Y.
Zhang
J.
Tang
X.
Qian
Y.
Gao
P.
Zhu
D.
Association of a functional single-nucleotide polymorphism in the ALDH2 gene with essential hypertension depends on drinking behavior in a Chinese Han population
J. Hum. Hypertens.
 
2013
27
181
186
30
Low
S.K.
Takahashi
A.
Cha
P.C.
Zembutsu
H.
Kamatani
N.
Kubo
M.
Nakamura
Y.
Genome-wide association study for intracranial aneurysm in the Japanese population identifies three candidate susceptible loci and a functional genetic variant at EDNRA
Hum. Mol. Genet.
 
2012
21
2102
2110
31
Tan
A.
Sun
J.
Xia
N.
Qin
X.
Hu
Y.
Zhang
S.
Tao
S.
Gao
Y.
Yang
X.
Zhang
H.
A genome-wide association and gene–environment interaction study for serum triglycerides levels in a healthy Chinese male population
Hum. Mol. Genet.
 
2012
21
1658
1664
32
Kamatani
Y.
Matsuda
K.
Okada
Y.
Kubo
M.
Hosono
N.
Daigo
Y.
Nakamura
Y.
Kamatani
N.
Genome-wide association study of hematological and biochemical traits in a Japanese population
Nat. Genet.
 
2010
42
210
215
33
Takeuchi
F.
Yokota
M.
Yamamoto
K.
Nakashima
E.
Katsuya
T.
Asano
H.
Isono
M.
Nabika
T.
Sugiyama
T.
Fujioka
A.
Genome-wide association study of coronary artery disease in the Japanese
Eur. J. Hum. Genet.
 
2012
20
333
340
34
Cui
R.
Kamatani
Y.
Takahashi
A.
Usami
M.
Hosono
N.
Kawaguchi
T.
Tsunoda
T.
Kamatani
N.
Kubo
M.
Nakamura
Y.
Functional variants in ADH1B and ALDH2 coupled with alcohol and smoking synergistically enhance esophageal cancer risk
Gastroenterology
 
2009
137
1768
1775
35
Adzhubei
I.A.
Schmidt
S.
Peshkin
L.
Ramensky
V.E.
Gerasimova
A.
Bork
P.
Kondrashov
A.S.
Sunyaev
S.R.
A method and server for predicting damaging missense mutations
Nat. Methods
 
2010
7
248
249
36
Kumar
P.
Henikoff
S.
Ng
P.C.
Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm
Nat. Protoc.
 
2009
4
1073
1081
37
Chen
L.
Davey
S.G.
Harbord
R.M.
Lewis
S.J.
Alcohol intake and blood pressure: a systematic review implementing a Mendelian randomization approach
PLoS Med.
 
2008
5
e52
38
Kim
Y.J.
Go
M.J.
Hu
C.
Hong
C.B.
Kim
Y.K.
Lee
J.Y.
Hwang
J.Y.
Oh
J.H.
Kim
D.J.
Kim
N.H.
Large-scale genome-wide association studies in East Asians identify new genetic loci influencing metabolic traits
Nat. Genet.
 
2011
43
990
995
39
Baik
I.
Cho
N.H.
Kim
S.H.
Han
B.G.
Shin
C.
Genome-wide association studies identify genetic loci related to alcohol consumption in Korean men
Am. J. Clin. Nutr.
 
2011
93
809
816
40
Veyrieras
J.B.
Kudaravalli
S.
Kim
S.Y.
Dermitzakis
E.T.
Gilad
Y.
Stephens
M.
Pritchard
J.K.
High-resolution mapping of expression-QTLs yields insight into human gene regulation
PLoS Genet.
 
2008
4
e1000214
41
Lee
J.Y.
Lee
B.S.
Shin
D.J.
Woo
P.K.
Shin
Y.A.
Joong
K.K.
Heo
L.
Young
L.J.
Kyoung
K.Y.
Jin
K.Y.
et al.  
A genome-wide association study of a coronary artery disease risk variant
J. Hum. Genet.
 
2013
58
120
126
42
Go
M.J.
Hwang
J.Y.
Kim
Y.J.
Hee
O.J.
Kim
Y.J.
Heon
K.S.
Soo
P.K.
Lee
J.
Kim
B.J.
Han
B.G.
et al.  
New susceptibility loci in MYL2, C12orf51 and OAS1 associated with 1-h plasma glucose as predisposing risk factors for type 2 diabetes in the Korean population
J. Hum. Genet.
 
2013
58
362
365
43
Cho
Y.S.
Go
M.J.
Kim
Y.J.
Heo
J.Y.
Oh
J.H.
Ban
H.J.
Yoon
D.
Lee
M.H.
Kim
D.J.
Park
M.
et al.  
A large-scale genome-wide association study of Asian populations uncovers genetic factors influencing eight quantitative traits
Nat. Genet.
 
2009
41
527
534
44
Cai
T.
Yu
P.
Monga
S.P.
Mishra
B.
Mishra
L.
Identification of mouse itih-4 encoding a glycoprotein with two EF-hand motifs from early embryonic liver
Biochim. Biophys. Acta
 
1998
1398
32
37
45
Choi
J.W.
Liu
H.
Choi
D.K.
Oh
T.S.
Mukherjee
R.
Yun
J.W.
Profiling of gender-specific rat plasma proteins associated with susceptibility or resistance to diet-induced obesity
J. Proteomics.
 
2012
75
1386
1400
46
Fujita
Y.
Ezura
Y.
Emi
M.
Sato
K.
Takada
D.
Iino
Y.
Katayama
Y.
Takahashi
K.
Kamimura
K.
Bujo
H.
Hypercholesterolemia associated with splice-junction variation of inter-a-trypsin inhibitor heavy chain 4 (ITIH4) gene
J. Hum. Genet.
 
2004
49
24
28
47
Zeller
T.
Wild
P.
Szymczak
S.
Rotival
M.
Schillert
A.
Castagne
R.
Maouche
S.
Germain
M.
Lackner
K.
Rossmann
H.
Genetics and beyond - the transcriptome of human monocytes and disease susceptibility
PLoS One
 
2010
5
e10693
48
Stranger
B.E.
Nica
A.C.
Forrest
M.S.
Dimas
A.
Bird
C.P.
Beazley
C.
Ingle
C.E.
Dunning
M.
Flicek
P.
Koller
D.
Population genomics of human gene expression
Nat. Genet.
 
2007
39
1217
1224
49
Montgomery
S.B.
Sammeth
M.
Gutierrez-Arcelus
M.
Lach
R.P.
Ingle
C.
Nisbett
J.
Guigo
R.
Dermitzakis
E.T.
Transcriptome genetics using second generation sequencing in a Caucasian population
Nature
 
2010
464
773
777
50
Xu
Z.
Taylor
J.A.
SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies
Nucleic Acids Res.
 
2009
37
W600
W605
51
Ripke
S.
Sanders
A.R.
Kendler
K.S.
Levinson
D.F.
Sklar
P.
Holmans
P.A.
Genome-wide association study identifies five new schizophrenia loci
Nat. Genet.
 
2011
43
969
976
52
Bergen
S.E.
O'Dushlaine
C.T.
Ripke
S.
Lee
P.H.
Ruderfer
D.M.
Akterin
S.
Moran
J.L.
Chambert
K.D.
Handsaker
R.E.
Backlund
L.
et al.  
Genome-wide association study in a Swedish population yields support for greater CNV and MHC involvement in schizophrenia compared with bipolar disorder
Mol. Psychiatry
 
2012
17
880
886
53
Smoller
J.W.
Craddock
N.
Kendler
K.
Lee
P.H.
Neale
B.M.
Nurnberger
J.I.
Ripke
S.
Santangelo
S.
Sullivan
P.F.
Identification of risk loci with shared effects on five major psychiatric disorders: a genome-wide analysis
Lancet
 
2013
381
1371
1379
54
Ehret
G.B.
Munroe
P.B.
Rice
K.M.
Bochud
M.
Johnson
A.D.
Chasman
D.I.
Smith
A.V.
Tobin
M.D.
Verwoert
G.C.
Hwang
S.J.
et al.  
Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk
Nature
 
2011
478
103
109
55
Wain
L.V.
Verwoert
G.C.
O'Reilly
P.F.
Shi
G.
Johnson
T.
Johnson
A.D.
Bochud
M.
Rice
K.M.
Henneman
P.
Smith
A.V.
et al.  
Genome-wide association study identifies six new loci influencing pulse pressure and mean arterial pressure
Nat. Genet.
 
2011
43
1005
1011
56
Newton-Cheh
C.
Johnson
T.
Gateva
V.
Tobin
M.D.
Bochud
M.
Coin
L.
Najjar
S.S.
Zhao
J.H.
Heath
S.C.
Eyheramendy
S.
et al.  
Genome-wide association study identifies eight loci associated with blood pressure
Nat. Genet.
 
2009
41
666
676
57
Hotta
K.
Kitamoto
A.
Kitamoto
T.
Mizusawa
S.
Teranishi
H.
Matsuo
T.
Nakata
Y.
Hyogo
H.
Ochi
H.
Nakamura
T.
et al.  
Genetic variations in the CYP17A1 and NT5C2 genes are associated with a reduction in visceral and subcutaneous fat areas in Japanese women
J. Hum. Genet.
 
2012
57
46
51
58
Meyer
B.
Wittig
I.
Trifilieff
E.
Karas
M.
Schagger
H.
Identification of two proteins associated with mammalian ATP synthase
Mol. Cell Proteomics.
 
2007
6
1690
1699
59
Dichgans
M.
Malik
R.
Konig
I.R.
Rosand
J.
Clarke
R.
Gretarsdottir
S.
Thorleifsson
G.
Mitchell
B.D.
Assimes
T.L.
Levi
C.
et al.  
Shared genetic susceptibility to ischemic stroke and coronary artery disease: a genome-wide analysis of common variants
Stroke
 
2014
45
24
36
60
Kamenecka
T.M.
Busby
S.A.
Kumar
N.
Choi
J.H.
Banks
A.S.
Vidovic
D.
Cameron
M.D.
Schurer
S.C.
Mercer
B.A.
Hodder
P.
et al.  
Potent anti-diabetic actions of a novel non-agonist PPARgamma ligand that blocks Cdk5-mediated phosphorylation. National Center for Biotechnology Information (US), Bethesda (MD)
2011
61
Choi
J.H.
Banks
A.S.
Estall
J.L.
Kajimura
S.
Bostrom
P.
Laznik
D.
Ruas
J.L.
Chalmers
M.J.
Kamenecka
T.M.
Bluher
M.
et al.  
Anti-diabetic drugs inhibit obesity-linked phosphorylation of PPARgamma by Cdk5
Nature
 
2010
466
451
456
62
Gragnoli
C.
Depression and type 2 diabetes: cortisol pathway implication and investigational needs
J. Cell Physiol.
 
2012
227
2318
2322
63
Cochran
W.G.
The combination of estimates from different experiments
Biometrics
 
1954
10
101
129
64
Devlin
B.
Roeder
K.
Genomic control for association studies
Biometrics
 
1999
55
997
1004

Author notes

Co-first authors.
Co-last authors.