A0826
Title: Two-sample Mendelian randomization for summary statistics accounting linkage disequilibrium
Authors: Qing Cheng - Duke-NUS (Singapore) [presenting]
Abstract: Obtaining a reliable causal relationship between risk exposures and disease outcomes from epidemiological studies remains an essential challenge. Proliferation of genome-wide association studies (GWAS) has prompted the use of two-sample Mendelian randomization (MR) with genetic variants as instrumental variables (IV) for data analysis and interpretation. However, most of existing methods assume that IVs are not in linkage disequilibrium(LD) which can lead to biased estimates and false-positive casual relationships. To overcome these limitations, we propose a probabilistic model thatleverages GWAS summary statistics in the presence of LD, as well as properly accounts for horizontal pleiotropy among genetic variants (MR-LDP). MR-LDP utilizes a computationally efficient variational Bayes expectation-maximization (VBEM) algorithm, calibrating evidence lower bound (ELBO) for a likelihood ratio test. We further conducted comprehensive simulation studies to demonstrate the advantages of MR-LDP over existing methods in terms of both type-I error control and point estimates. Moreover, we used two real expsoure-outcome pairs to validate results from MR-LDP in comparison with alternative methods, particularly showing our method is more efficient using all genetic variants in LD.
