JointPRS: A data-adaptive framework for multi-population genetic risk prediction incorporating genetic correlation

• DOI: 10.1038/s41467-025-59243-x
• Publication Date: 2025-04-24T01:39:01Z
• Authors: Leqi Xu; Geyu Zhou; Wei Jiang; Haoyu Zhang; Yikai Dong; Leying Guan; Hongyu Zhao
• Keywords: Genetic data; Genetic correlation

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 using GWAS summary statistics. It achieves accurate predictions without individual-level data remains effective in the presence of set thanks to its approach. Through extensive simulations real applications 22 quantitative four binary traits five continental evaluated UK Biobank (UKBB) All Us (AoU), JointPRS consistently outperforms six state-of-the-art methods three scenarios: no data, same-cohort testing, cross-cohort testing. Notably, Admixed American population, improves lipid trait AoU 6.46%-172.00% compared other existing methods.