Back

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

Yang, Y.; Zhu, X.

2025-08-08 bioinformatics
10.1101/2025.07.11.664423 bioRxiv
Show 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.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.