Resampling-Based Variable Selection Methods for Next-Generation Sequencing Studies

• Activity Number: 164
• Type: Topic Contributed
• Date/Time: Monday, July 30, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Michael C Wu*+
• Companies: The University of North Carolina at Chapel Hill
• Keywords: Next generation sequencing studies ; High-dimensional data ; Variable selection ; Stability selection ; Genetic association studies

Abstract

The development of next-generation sequencing technology has resulted in studies for identifying rare genetic variants associated with complex traits. Since statistical methods for common genetic variants are underpowered for testing the effect of individual rare variants, region based analysis, wherein multiple variants in a region are grouped and their cumulative effect on the trait is tested, has become the standard approach for analyzing rare variants. Although these methods are able to identify groups of variants that are associated with traits, they cannot identify individual causal variants. In order to select individual variants, we propose a simple resampling based variable selection method that closely mimics and reflects the principles underlying stability selection yet is tailored toward rare variant analysis. We use simulations and real data analyses to demonstrate the power of our approach.