Multi-ancestry

Data-adaptive polygenic score framework that borrows strength across populations via genetic correlations using only GWAS summary statistics and LD references—supporting prediction with or without individual-level tuning data.

PRS, multi-population, genetic correlation, summary statistics, cross-ancestry

Genetic risk prediction for non-European populations is hindered by limited Genome-Wide Association Study (GWAS) sample sizes and small tuning datasets. We propose JointPRS, a data-adaptive framework that leverages genetic correlations across multiple populations using GWAS summary statistics. It achieves accurate predictions without individual-level tuning data and remains effective in the presence of a small tuning set thanks to its data-adaptive approach. Through extensive simulations and real data applications to 22 quantitative and four binary traits in five continental populations evaluated using the UK Biobank (UKBB) and All of Us (AoU), JointPRS consistently outperforms six state-of-the-art methods across three data scenarios: no tuning data, same-cohort tuning and testing, and cross-cohort tuning and testing. Notably, in the Admixed American population, JointPRS improves lipid trait prediction in AoU by 6.46%-172.00% compared to the other existing methods.

gnomAD v4 dramatically expanded to 807,098 exomes and 76,215 genomes, with substantially improved ancestral diversity. v4 included >30 genetic ancestry groups, enabling more precise population-specific allele frequency estimates. New constraint metrics across diverse ancestries and improved structural variant calls made this the most comprehensive human variation resource ever created.

Freeze 8 expanded WGS to ~62k participants, substantially increasing representation of African American, Hispanic/Latino, Asian, and other ancestries. Enabled deep variant discovery with >400M variants including many rare and population-specific alleles. Became the gold-standard imputation reference panel for multi-ethnic studies.

Multi-ancestry GWAS meta-analysis across 7,266 traits in UK Biobank, identifying 14,676 significant loci not found in EUR-only analysis.