Improving causal effect estimation in multi-ancestry multivariable Mendelian randomization with transfer learning

Abstract

Multivariable Mendelian randomization (MVMR) has been largely limited to individuals of European ancestry, due to the larger sample sizes available in European genome-wide association studies (GWAS). We introduce MRBEE-TL, one of the first multi-ancestry MVMR methods, which combines transfer learning with bias-corrected estimating equations to improve power in underpowered ancestries and to assess cross-ancestry heterogeneity of disease risk factors. In simulations, MRBEE-TL consistently outperformed MR methods that relied solely on ancestry-specific GWAS data. In real data analyses, MRBEE-TL not only identified ancestry-consistent and ancestry-specific causal effects missed by conventional methods, but also improved power in African and East Asian ancestries. MRBEE-TL is available through the R package MRBEEX at https://github.com/harryyiheyang/MRBEEX.