Variable Selection for Genome-Wide Multiple Loci Mapping

• Activity Number: 411
• Type: Topic Contributed
• Date/Time: Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Wei Sun*+ and Fei Zou and Joseph G. Ibrahim
• Companies: The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
• Keywords: adaptive lasso ; varaible selection ; multiple loci mapping ; QTL ; association ; genetics

Abstract

Despite many recent methodology developments, variable selection in high dimensional setting remains a difficult problem, especially when the covariates with zero coefficients are correlated with some covariates with nonzero coefficients. One such example is genome-wide multiple loci mapping with dense genetic markers. The adaptive Lasso (Zou, H. 2006) is a state-of-art method for simultaneous variable selection and estimation in the setting of linear regression. However, it requires consistent initial estimates of the regression coefficients, which are generally not available in the aforementioned high-dimensional setting. We propose two variable selection methods for multiple loci mapping. These methods extend the adaptive Lasso in the sense that they do not require any initial estimates of the regression coefficients.