We conducted the largest trans-ethnic meta-analysis of genome-wide association studies of type 2 diabetes (T2D) by aggregating 171,262 cases and 1,075,072 controls from diverse populations (doubling the effective sample size from recently published European-specific analysis). We identified 250 loci associated with T2D at genome-wide significance (p<5x10-8; 59 novel), which were delineated to 475 distinct signals (p<10-5). Allelic effects on T2D at index variants for most signals were homogenous across populations: only 17% displayed evidence of effect-size heterogeneity related to ancestry (pHET <1.4x10-4, Bonferroni correction). T2D-associated variants were jointly enriched (p<0.00023, correction for 220 annotations) in coding exons, pancreatic islet enhancers and promoters, adipose enhancers, liver enhancers, and binding sites for transcription factors (NKX2.2, PDX1, FOXA2, EZH2). Compared to previous efforts, increased sample size, population diversity, and annotation-informed fine-mapping appreciably improved localisation of causal variants achieving single variant resolution at 217 signals (>50% of the posterior probability of association assigned to a single variant). Through co-localisation of these high-confidence causal variants with protein and RNA QTLs (in disease-relevant tissues), and chromatin conformation capture data (in human islets), we identified >80 candidate effector genes including several not previously implicated e.g., CARD9, PDE8B, ABCB10, SIX3, INHBB. Genetic risk scores derived from ethnic-specific effect sizes at lead variants were transferrable across populations, with minimum liability-scale variance explained from Europeans into Africans. T2D-associated variants displayed evidence of directional adaptation through time, indicating selection for increased T2D risk in African populations, relative to elsewhere in the world. These analyses represent the most comprehensive view of the genetic contribution to T2D to date.


A. Mahajan: None. X. Sim: None. W. Zhang: None. J.E. Below: None. H. Kitajima: None. L. Speidel: None. A. Morris: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.