Recent positive selection implicates <i>IP6K3</i> and <i>MAPT</i> as metabolically relevant loci in South Asians

Abstract South Asians constitute 25% of the global population yet account for 33% of individuals living with type 2 diabetes (T2D) (1). Compared with individuals of European ancestry, South Asians develop metabolic disease at lower body mass index and progress more rapidly to cardiometabolic complication (2,3). Despite disproportionate disease burden, South Asians represented less than 1% of participants in genome-wide association studies (4). This underrepresentation limits biological insight and hampers genomic discovery. Here we integrate genome-wide signatures of recent positive selection with cross-trait genetic association data across 13 South Asian populations (5,6) (n = 676 whole genomes). We identify 1,797 genes putatively under selection, prioritise 65 shared across the subcontinent, and refine four loci— RBM6 , IP6K3 , MAPT and PEPD —using colocalisation (from publicly available multi-ancestry summary statistics) and multi-trait fine-mapping (7,8) in two independent south Asian cohorts (9,10). Integrating signatures of recent positive selection with genetic association data prioritises metabolically relevant loci for T2D in South Asians and establishes a scalable framework for locus discovery in underrepresented populations, without reliance on ancestry-matched molecular quantitative trait loci (QTL) resources.