Comment on: Chauhan et al. (2010) Impact of common variants of PPARG, KCNJ11, TCF7L2, SLC30A8, HHEX, CDKN2A, IGF2BP2, and CDKAL1 on the risk of type 2 diabetes in 5,164 Indians. Diabetes;59:2068-2074.

In the last 4 years, genome-wide association studies (GWASs) have revolutionized the complex disease genomics research worldwide. In the context of type 2 diabetes, GWASs have led to the discovery of several novel genetic markers that illuminated hitherto unknown biological pathways. Now, the validation of these type 2 diabetes–related priority single nucleotide polymorphisms (SNPs) (from GWASs) among different world populations is needed. Despite a very large burden of type 2 diabetes in India and the important question of possible genetic susceptibility among south Asians, the lack of GWASs and the small number of candidate gene studies from the Indian subcontinent are disappointing. Therefore, the validation study of Chauhan et al. (1) of eight GWAS-identified SNPs on a large Indian sample ( 5,000) that demonstrated strong evidence of association for all eight SNPs with type 2 diabetes is a very important contribution to the literature. However, despite a large sample size, the reliability of these results may be questioned because confounding due to population stratification was not formally assessed. Detection of hidden stratification requires statistical tools (e.g., genomic control, structure analysis, principal component analysis) using a set of unlinked genetic markers. In their study, Chauhan et al. (1) considered that population stratification was unlikely because both case subjects and control subjects were recruited from same place (i.e., Delhi and Pune). Both places are melting pots of various endogamous (inter-marriage within clans) populations, and the chances of hidden stratification may be high. The landmark study of Reich et al. (2) clearly showed the highly stratified structure of the Indian population. They conducted the genome-wide survey on 25 Indian populations and confirmed the population heterogeneity and demonstrated the endogamous nature of the Indian population. A similar conclusion was made by the Indian Genome Variation Consortium (3) by genotyping 405 SNPs in 55 populations (1,871 individuals). Consideration of recruitment strategies based on endogamy in a defined geographic area would be one means of conducting genetic association studies in India without introducing population stratification. The results of Chauhan et al. are highly plausible given the high priors for these SNPs from studies in other populations, but the possibility of population stratification, particularly when evaluating future genetic markers for which there are no strong priors, should be taken into account and appropriate analyses conducted to demonstrate that the findings are robust.