Multi-ancestry genetic study of type 2 diabetes highlights the power of diverse populations for discovery and translation

Abstract

We assembled an ancestrally diverse collection of genome-wide association studies (GWAS) of type 2 diabetes (T2D) in 180,834 affected individuals and 1,159,055 controls (48.9% non-European descent) through the Diabetes Meta-Analysis of Trans-Ethnic association studies (DIAMANTE) Consortium. Multi-ancestry GWAS meta-analysis identified 237 loci attaining stringent genome-wide significance (P < 5 × 10^-9), which were delineated to 338 distinct association signals. Fine-mapping of these signals was enhanced by the increased sample size and expanded population diversity of the multi-ancestry meta-analysis, which localized 54.4% of T2D associations to a single variant with >50% posterior probability. This improved fine-mapping enabled systematic assessment of candidate causal genes and molecular mechanisms through which T2D associations are mediated, laying the foundations for functional investigations. Multi-ancestry genetic risk scores enhanced transferability of T2D prediction across diverse populations. Our study provides a step toward more effective clinical translation of T2D GWAS to improve global health for all, irrespective of genetic background.

Extended Data Fig. 1. Study overview.

Summary of data resources and downstream analyses to identify candidate causal genes at T2D susceptibility loci.

Extended Data Fig. 2. Axes of genetic variation separating GWAS of T2D across diverse populations.

The first three axes of genetic variation (PC 1, PC 2 and PC 3) from multi-dimensional scaling of the Euclidean distance matrix between populations are sufficient to separate five ancestry groups: African (AFR), East Asian (EAS), European (EUR), Hispanic (HIS) and South Asian (SAS). The second axis of genetic variation (PC 2) separates African American and continental African GWAS. The third axis of genetic variation (PC 3) reveals finer-scale differences between GWAS within ancestry groups: Hispanic studies with a greater proportion of American ancestry (SIGMA (2), MC (1) and MC (2)) or African ancestry (WHI, MESA, HCHS/SOL and BIOME); East Asian studies of Chinese, Japanese and Korean ancestry from those of Malay and Filipino ancestry (SIMES and CLHNS); South Asian studies of Sri Lankan, Bangladeshi and South Indian ancestry (RHS, EPIDREAM, SINDI, GRCCDS and BPC) from those of North Indian and Pakistani ancestry; and Northern European ancestry studies from the study of Greek ancestry from Southern Europe (GOMAP). GWAS were aligned to ancestry groups based on self-report at the study level.

Extended Data Fig. 3. Manhattan plot of genome-wide T2D association from multi-ancestry meta-regression (MR-MEGA) of up to 180,834 cases and 1,159,055 controls.

Each point represents an SNV passing quality control in the multi-ancestry meta-regression, plotted with their association P-value (on a −log₁₀ scale, truncated at 300) as a function of genomic position (NCBI build 37). Association signals attaining genome-wide significance are highlighted in pale blue (P < 5 x 10⁻⁹) and dark blue (P < 5 x 10⁻⁸). The names of novel loci names are highlighted with their association P-value from the multi-ancestry meta-regression.

Extended Data Fig. 4. Comparison of association P-values at lead SNVs at T2D loci between multi-ancestry meta-regression (MR-MEGA), fixed-effects meta-analysis and random-effects (RE2) meta-analysis of up to 180,834 cases and 1,159,055 controls.

Each point corresponds to an SNV, plotted according to P-values (on a −log₁₀ scale) from MR-MEGA on the x-axis and fixed- or random-effects meta-analysis on the y-axis. SNVs below the y = x line demonstrate stronger association with MR-MEGA. The lead SNV at the TCF7L2 locus has been removed to improve clarity of presentation.

Extended Data Fig. 5. Comparison of loci identified at genome-wide significance (P < 5 x 10⁻⁸) in multi-ancestry meta-regression (180,834 cases and 1,159,055 controls), and East Asian and European ancestry-specific meta-analyses (56,268 cases and 227,155 controls, and 80,154 cases and 853,816 controls, respectively).

a, Association P-values at loci identified in East Asian and European ancestry-specific meta-analyses. Each point corresponds to a locus, plotted according to the P-value (on a −log₁₀ scale) for the lead SNP in the multi-ancestry meta-regression on the x-axis and the lead SNP in the ancestry-specific meta-analysis on the y-axis. The TCF7L2 locus has been removed to improve clarity of presentation. Loci plotted below the y = x line show stronger evidence for association in the multi-ancestry meta-regression. b, Overlap of loci identified in multi-ancestry meta-regression and ancestry-specific meta-analyses.

Extended Data Fig. 6. Summary statistics from joint fGWAS model of enriched functional and regulatory annotations across distinct T2D association signals from multi-ancestry meta-regression (MR-MEGA) of up to 180,834 cases and 1,159,055 controls.

Each point corresponds to an annotation, plotted for the log-enrichment for T2D association on the x-axis, with bars representing the corresponding 95% confidence interval (CI).

Extended Data Fig. 7. Comparison of number of SNVs in 99% credible set for distinct association signals for T2D obtained from the multi-ancestry meta-regression of 180,834 cases and 1,159,055 controls under uniform and annotation-informed prior models of causality.

Each point corresponds to a distinct association signal, plotted according to the log₁₀ credible set size under the uniform prior on the x-axis and the log₁₀ credible set size under the annotation-informed prior on the y-axis. The 144 (42.6%) signals below the y = x line were more precisely fine-mapped under the annotation-informed prior.

Extended Data Fig. 8. Differences in LD structure between ancestry groups at the PROX1 locus for distinct association signals from multi-ancestry meta-regression (MR-MEGA) of up to 180,840 cases and 1,159,185 controls.

Each point represents an SNV passing quality control in the multi-ancestry meta-regression (after conditional analysis), plotted with their association P-value (on a log₁₀ scale) as a function of genomic position (NCBI build 37). The index SNV is represented by the purple symbol. The color coding of all other SNVs indicates LD with the index variant in the ancestry-matched reference haplotypes from the 1000 Genomes Project panel: red, r² ≥ 0.8; gold, 0.6 ≤ r² < 0.8; green, 0.4 ≤ r² < 0.6; cyan, 0.2 ≤ r² < 0.4; blue, r² < 0.2; grey, r² unknown. Recombination rates are estimated from Phase II HapMap and gene annotations are taken from the University of California Santa Cruz genome browser.

Extended Data Fig. 9. Power of multi-ancestry GRS to predict T2D status in 129,230 individuals of Finnish ancestry from FinnGen.

a, Age under receiver operating characteristic curve (AUROC) after adding BMI and GRS to a baseline model adjusting for age and sex. b, Prevalence of T2D across GRS deciles. c, Boxplot of the distribution of age at T2D diagnosis across GRS deciles: box defines upper quartile, median and lower quartile, bars define maximum and minimum values within 1.5 x interquartile range of the upper and lower quartiles, other points are outliers.

Extended Data Fig. 10. Evidence for selection from Relate in African ancestry populations of subsets of T2D risk variants (effect aligned to derived allele) that are associated with other traits available in the UK Biobank.

Nominal evidence for selection (P < 0 .05) is indicated by the dashed line. The color of each point indicates the evidence for selection of subsets of T2D risk variants that are not associated with the other trait: P < 0.05 (pink) and P ≥ 0.05 (black). Population abbreviations: ESN, Esan in Nigeria; GWD, Gambian in Western Divisions in the Gambia; LWK, Luhya in Webuye, Kenya; MSL, Mende in Sierra Leone; YRI, Yoruba in Ibadan, Nigeria.

Figure 1 ∣. Comparison of fine-mapping resolution for distinct association signals for T2D obtained from ancestry-specific meta-analysis and multi-ancestry meta-regression.

a, Each point corresponds to a distinct association signal, plotted according to the log₁₀ credible set size in the multi-ancestry meta-regression on the x-axis and the log₁₀ credible set size in the European ancestry meta-analysis on the y-axis. The 266 (78.7%) signals above the dashed y = x line were more precisely fine-mapped in the multi-ancestry meta-regression. b, We “down-sampled” the multi-ancestry meta-regression to the effective sample size of the European ancestry-specific meta-analysis. Each point corresponds to one of the 266 signals that were more precisely fine-mapped in the multi-ancestry meta-regression. The 137 (51.5%) signals above the dashed y = x line were more precisely fine-mapped in “down-sampled” multi-ancestry meta-regression than the equivalent sized European ancestry-specific meta-analysis. c, Properties of 99% credible sets of variants driving each distinct association signal in European ancestry-specific meta-analysis, combined East Asian and European ancestry meta-analysis, and multi-ancestry meta-regression. The inclusion of the most under-represented ancestry groups (African, Hispanic and South Asian) in the multi-ancestry meta-regression reduced the median size of 99% credible sets and increased the median posterior probability ascribed to index SNVs.

Figure 2 ∣. T2D association signal at the BCAR1 locus colocalizes with multiple circulating plasma pQTLs.

a, Signal plot for T2D association from multi-ancestry meta-regression of 180,834 cases and 1,159,055 controls of diverse ancestry. Each point represents an SNV, plotted with their P-value (on a log₁₀ scale) as a function of genomic position (NCBI build 37). Gene annotations are taken from the University of California Santa Cruz genome browser. Recombination rates are estimated from the Phase II HapMap. b, Fine-mapping of T2D association signal from multi-ancestry meta-regression. Each point represents an SNV plotted with their posterior probability of driving T2D association as a function of genomic position (NCBI build 37). Chromatin states are presented for four diabetes-relevant tissues: active TSS (red), flanking active TSS (orange red), strong transcription (green), weak transcription (dark green), genic enhancers (green yellow), active enhancer (orange), weak enhancer (yellow), bivalent/poised TSS (Indian red), flanking bivalent TSS/enhancer (dark salmon), repressed polycomb (silver), weak repressed polycomb (Gainsboro), quiescent/low (white). c, Schematic presentation of the single cis- and multiple trans- effects mediated by the BCAR1 locus on plasma proteins and the islet chromatin loop between islet enhancer and promoter elements near CTRB2. d, Signal plots for four circulating plasma proteins that colocalize with the T2D association in 3,301 European ancestry participants from the INTERVAL study. Each point represents an SNV, plotted with their P-value (on a log₁₀ scale) as a function of genomic position (NCBI build 37). e, Expression of genes (transcripts per million, TPM) encoding colocalized proteins in islets, pancreas and whole blood.

Figure 3 ∣. Defining causal molecular mechanisms at the PROX1 locus.

a, Signal plot for two distinct T2D associations from multi-ancestry meta-regression of 180,834 cases and 1,159,055 controls of diverse ancestry. Each point represents an SNV, plotted with their P-value (on a −log₁₀ scale) as a function of genomic position (NCBI build 37). Index SNVs are represented by the blue and purples diamonds. All other SNVs are colored according to the LD with the index SNVs in European and East Asian ancestry populations. Gene annotations are taken from the University of California Santa Cruz genome browser. b, Fine-mapping of T2D association signals from multi-ancestry meta-regression. Each point represents a SNV plotted with their posterior probability of driving each distinct T2D association as a function of genomic position (NCBI build 37). The 99% credible sets for the two signals are highlighted by the purple and blue diamonds. Chromatin states are presented for four diabetes-relevant tissues: active TSS (red), flanking active TSS (orange red), strong transcription (green), weak transcription (dark green), genic enhancers (green yellow), active enhancer (orange), weak enhancer (yellow), bivalent/poised TSS (Indian red), flanking bivalent TSS/enhancer (dark salmon), repressed polycomb (silver), weak repressed polycomb (Gainsboro), quiescent/low (white). c, Transcriptional activity of the 99 credible set variants at the two T2D association signals in human HepG2 hepatocytes and EndoC-βH1 beta cell models obtained from in vitro reporter assays. Biological replicates: n = 3. Technical replicates: n = 3. WT, wild-type (non-risk allele/haplotype); GFP, green fluorescent protein (negative control); EV, empty vector (baseline). Height of bar represents mean. Error bars represent standard error of the mean. Differences in luciferase activity between groups were tested using two-tailed two-sample t-tests, where P < 0.05 was considered statistically significant. d, Expression of PROX1 (transcripts per million, TPM) across a range of diabetes-relevant tissues.

Figure 4 ∣. Transferability of multi-ancestry and ancestry-specific GRS into GWAS across diverse population groups.

Each GRS was constructed using lead SNVs attaining genome-wide significance (P < 5 x 10⁻⁹ for multi-ancestry GRS and P < 5 x 10⁻⁸ for ancestry-specific GRS). For the multi-ancestry GRS, population-specific allelic effects on T2D were estimated from the meta-regression to generate different GRS weights for each test GWAS. For each ancestry-specific GRS, weights were generated from allelic effect estimates obtained from fixed-effects meta-analysis. a, The trait variance explained (pseudo R²) by each GRS was assessed in two test GWAS from each ancestry group. b, The multi-ancestry GRS out-performed ancestry-specific GRS into all test GWAS, reflecting the shared genetic contribution to T2D across diverse populations, despite differing allele frequencies and LD patterns.

Figure 5 ∣. Positive selection acting on T2D index SNVs.

a, Evidence of selection from Relate towards increased T2D risk is restricted to African ancestry populations and is explained by those SNVs that are associated with increased weight. b, T2D risk alleles that are associated with increased weight are particularly young for their derived allele frequency (DAF). Population abbreviations (sample sizes): ESN (98), Esan in Nigeria; GWD (112), Gambian in Western Divisions of the Gambia; LWK (98), Luhya in Webuye, Kenya; MSL (84), Mende in Sierra Leone; YRI (107), Yoruba in Ibadan, Nigeria; BEB (85), Bengali in Bangladesh; GIH (102), Gujarati Indian from Houston, Texas; ITU (101), Indian Telegu from the UK; PJL (95), Punjabi from Lahore, Pakistan; STU (101), Sri Lankan Tamil from the UK; CDX (92), Chinese Dai in Xishuangbanna, China; CHB (102), Han Chinese in Beijing, China; CHS (104), Southern Han Chinese; JPT (103), Japanese in Tokyo, Japan; KHV (98), Kinh in Ho Chi Min City, Vietnam; CEU (98), Utah residents with Northern and Western European ancestry; FIN (98), Finnish in Finland; GBR (90), British in England and Scotland; IBS (106), Iberian population in Spain; TSI (106), Toscani in Italy.