| Title: | Meta-evolutionary exome analysis identifies novel type 2 diabetes mellitus genes in the UK Biobank and all of us |
| Journal: | PLOS Genetics |
| Published: | 30 Sep 2025 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/41026804/ |
| DOI: | https://doi.org/10.1371/journal.pgen.1011889 |
Publication 19340
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Type 2 diabetes mellitus (T2DM) risk is heavily influenced by genetics, yet current association tests have explained only parts of its heritability. We developed MEVA (Meta-Evolutionary Action), a meta-analytic framework that integrates three complementary methods-EAML, Sigma-Diff, and GeneEMBED-to assess the functional burden of protein-coding variants using evolutionary data. MEVA was applied to exome data from 28,115 T2DM cases and 28,115 controls in the UK Biobank (UKB), identifying 101 genes (p < 1e-5). MEVA outperformed its component methods, each of which substantially outperformed a conventional burden test (MAGMA), in recovering known T2DM genes (AUROC = 0.925) and maintaining robustness in progressively smaller cohorts (AUROC = 0.917). MEVA showed significant enrichment for T2DM-related loci (p = 6.8e-10, p = 2.0e-34), protein interactions (z = 4.6, z = 4.2), pathways (p = 1.3e-6, z = 2.0), phenotypes (p = 1.3e-21, z = 9.1), and literature mentions (z = 7.2). Replication in 16,915 T2DM cases and 16,915 controls from All of Us (AoU) yielded 99 genes (p < 1e-5), 23 of which were also recovered in the UKB cohort - far exceeding random chance. These included established genes (SLC30A8, WFS1, HNF1A) and less-characterized candidates (NRIP1, ADAM30, CALCOCO2, TUBB1, ZFP36L2, WDR90). Notably, NRIP1 loss-of-function variants were associated with increased T2DM risk in both the UKB (OR = 1.09, FDR = 5.4e-4) and AoU (OR = 1.09, FDR = 0.046), and TUBB1 and CALCOCO2 gain-of-function variants showed consistent risk effects (FDR < 0.05). Pathway analyses revealed convergence on endoplasmic reticulum chaperone complexes (FDR = 0.02) and Hippo signaling (FDR = 8.5e-4). Finally, all 177 candidate genes were functionally prioritized using ten orthogonal criteria to guide experimental follow-up. These results demonstrate that combining complementary, impact-aware association tests increases sensitivity, improves replication, and expands the catalog of genetic risk factors for T2DM.</p>
12 Keywords
- Biological Specimen Banks
- Diabetes Mellitus, Type 2
- Evolution, Molecular
- Exome
- Exome Sequencing
- Female
- Genetic Predisposition to Disease
- Genome-Wide Association Study
- Humans
- Male
- UK Biobank
- United Kingdom
7 Authors
- Kevin Wilhelm
- Jennifer Asmussen
- Kwanghyuk Lee
- Maryam Samieinasab
- Emmanuel Asante
- Marek Kimmel
- Olivier Lichtarge
1 Application
| Application ID | Title |
|---|---|
| 55532 | Integrating the impact of exome variations |
Enabling scientific discoveries that improve human health