Bayesian LASSO for Genome-Wide Association Analysis

• Activity Number: 242
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Jiahan Li* and Rongling Wu+
• Companies: Penn State and Penn State
• Keywords: Bayesian ; Model Selection ; GWAS ; MCMC

Abstract

Recently, genome-wide association studies have successfully identified genes that may affect complex traits or diseases. Although standard statistical tests for each single-nucleotide polymorphism (SNP) separately are able to capture main genetic effects, different approaches are necessary to identify multiple SNPs that influence disease risk jointly or SNP-SNP interactions. In this paper, we present a Bayesian approach which simultaneously estimates all possible genetic effects associated with all SNPs. The proposed model considers a double-exponential prior for the variances, which results in a Bayesian LASSO model that is advantageous when the number of predictors far exceeds the number of observations. By doing this, the most relevant disease genes can be identified from a huge number of SNPs. The methods are illustrated using both simulated and real data.