A0247
Title: A Bayesian variable selection approach incorporating prior feature ordering and population structures
Authors: Thierry Chekouo - University of Minnesota (United States) [presenting]
Abstract: A novel Bayesian variable selection framework is proposed for the identification of important genetic variants associated with Coronary artery disease status. Instead of treating each feature independently as in conventional Bayesian variable selection methods, we propose an innovative prior for the inclusion probabilities of genetic variants that account for their ordering structure. We assume that neighboring variants are more likely to be selected together as they tend to be highly correlated and have similar biological functions. Additionally, we propose to group participating subjects based on underlying population structure and fit separate regressions, so that the regression coefficients can better reflect different disease risks in different population groups. Our approach borrows strength across regression models through an innovative prior inspired by the Markov random fields. The proposed framework can improve variable selection and prediction performances, as demonstrated in the simulation studies. We also apply the proposed framework to the CATHGEN data with binary CAD disease status.
