Identifying Gene-environment interactions with robust marginal Bayesian variable selection

TL;DR

This paper introduces a robust Bayesian variable selection method for gene-environment interaction studies that effectively handles outliers and data contamination, improving detection of meaningful genetic and environmental effects.

Contribution

It presents a novel marginal Bayesian approach with spike-and-slab priors for G×E studies, addressing robustness and outperforming existing methods in simulations and real data.

Findings

Method outperforms alternatives in simulations

Identifies biologically relevant gene-environment effects

Robust to outliers and data contamination

Abstract

In high-throughput genetics studies, an important aim is to identify gene-environment interactions associated with the clinical outcomes. Recently, multiple marginal penalization methods have been developed and shown to be effective in G$\times$E studies. However, within the Bayesian framework, marginal variable selection has not received much attention. In this study, we propose a novel marginal Bayesian variable selection method for G$\times$E studies. In particular, our marginal Bayesian method is robust to data contamination and outliers in the outcome variables. With the incorporation of spike-and-slab priors, we have implemented the Gibbs sampler based on MCMC. The proposed method outperforms a number of alternatives in extensive simulation studies. The utility of the marginal robust Bayesian variable selection method has been further demonstrated in the case studies using data from the Nurse Health Study (NHS). Some of the identified main and interaction effects from the real data analysis have important biological implications.