A new era for Type 2 diabetes genetics

The field of complex disease genetics has witnessed rapid progress over the past few months. The advent of genome-wide association scans (GWAS) has mediated a change in gear and the Type 2 diabetes (T2D) research community has set an unprecedented record with five genome-wide association studies published since February 2007 [1–6], increasing the number of confirmed Type 2 diabetes susceptibility loci from three ( PPARG , KCNJ11 , TCF7L2 ) to nine (with the addition of CDKAL1 , CDKN2A / B , IGF2BP2 , HHEX / IDE , FTO and SLC30A8 ). These studies have increased our understanding of the genetic aetiology of T2D and provided invaluable insights into the way genetic studies should be conducted. This review aims to summarize the major findings distilled from the dawn of this new era in T2D genetics. The feasibility of carrying out GWAS has been eagerly anticipated in the field of T2D genetics for some time. Several years of mostly unsuccessful gene hunting led to the realization that carefully designed and conducted, large-scale, biology-agnostic studies would be a necessary addition to the geneticist’s arsenal. Whole genome linkage scans, candidate gene-focused association studies and region-specific finemapping approaches had dominated the literature over the past few years. However, only three genes ( PPARG , KCNJ11 and, more recently, TCF7L2 ) had been established as T2D susceptibility loci [7–9]. Several reasons can account for the paucity of success stories in the scientific literature, but the biggest culprits stand out as insufficient power, over-interpretation of results and candidacy assessments based on incomplete knowledge of T2D aetiopathology. Lack of power in genetic association studies could be ascribed to a variety of factors, including insufficient sample sizes to detect the modest effects conferred by susceptibility variants, as well as an incomplete understanding of human genetic variation. Indeed, it was not unusual for gene mapping studies to examine only a handful of variants, or even a single variant, per gene (most likely in a coding portion of the transcribed unit).