Online Program Home
My Program

Abstract Details

Activity Number: 239 - Omics II
Type: Contributed
Date/Time: Monday, July 30, 2018 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #330929
Title: A Bilinear Regression Approach to Inform Variable Selection by Continuous Functional Annotation Information
Author(s): Pixu Shi* and Sunduz Keles and Ming Yuan
Companies: and University of Wisconsin, Madison and Columbia University
Keywords: Functional annotation; GWAS; Metagenomics; Microbiome; variable selection

In genome-wide association studies and metagenomic studies, the selection of important genotype features is often hindered by the strong association across features due to intrinsic biological reasons. This paper considers a high-dimensional bilinear regression approach that integrates continuous functional annotation information of genotype in the variable selection. The bilinear regression model can also provide an estimate of importance of each piece of auxiliary information. Simulation analysis shows advantage of the proposed method compared to uninformed variable selection, variable selection with discretized annotations and univariate regression. The proposed model is then applied to a genome-wide association study with histone modification annotations, and a microbiome metagenomic study with KEGG orthology annotations.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program