A Gene-Based Supervised Dimension Reduction Approach to Identify Common and Rare Variants in Genome-Wide Association Studies

• Activity Number: 233
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Asuman Turkmen*+ and Shili Lin
• Companies: The Ohio State University at Newark and The Ohio State University
• Keywords: genome-wide association studies ; rare variants ; dimension reduction ; PLS ; PCA

Abstract

Genome-wide association studies (GWAS) have become a popular approach for the identification of genes and genetic variants involved in complex diseases. While the standard approach for GWAS has been single-SNP analysis, multi-marker association methods in which multiple markers analyzed jointly to utilize more information have shown to provide more insight into association studies and lead to greater potential in identifying rare associated variants. In this study, we propose to aggregate the signals of many SNPs within a gene using latent components derived by a supervised dimension reduction method by which possible genetic effects related to rare variants can be revealed. Then, a penalized regression model is employed to relate the resulting latent components that represent the genes in the study and the trait of interest to detect associations. The performance of the proposed method is compared with currently available methods.