In the last decade, genome-wide association studies (GWAS) have revealed about hundreds of loci associated with type 2 diabetes(T2D) and related traits. However, most of loci have been discovered based on Europeans and a few GWAS conducted in populations with East Asian ancestry. To identify genetic variants responsible for complex traits such as T2D, obesity, and blood biochemical traits in Koreans, Korea Biobank Array(KBA) project was initiated in 2014. KBA is designed to contain about 830K tagging and functional variants optimized for Korean genome. In this study, quality controlled genotype data of about 35,000 samples was used. Among them, we conducted genome-wide association scan in 3,022 T2D case subjects and 27,195 control subjects. As a result, we replicated five previously known loci (CDKAL1, SLC30A8, CDKN2A, KCNQ1, and HNF1B) at genome-wide significance (P <= 5x10-8). We further compared genetic effects in Koreans with those of previously reported loci. As of March 2018, there were 373 unique variants of 50 T2D GWAS papers listed in GWAS catalogue. Among 373 variants, 359 variants were available after excluding variants with low imputation quality (<0.3). We compared odds ratios of 359 variants between KBA T2D GWAS and previously reported results. In overall, Pearson correlation coefficient of odds ratios was -0.04. For replicated results (130 variants with P<0.05 in KBA T2D GWAS), however, we observed high correlation (r<0.58). Our study identified five known T2D loci and confirmed consistent genetic effects across populations for 130 previously reported T2D associated variants. Further study is warranted to examine genetic diversity and lack of statistical power due to insufficient sample size in non-replicated loci. These results will be valuable scientific evidence in T2D precision medicine considering genetic effects across populations.
Y. Kim: None. M. Hwang: None. D. Shin: None.