|
Activity Number:
|
622
- Statistical Methods for Genome- and Epigenome-Wide Association Studies and Gene Environment Interactions
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #307235
|
Presentation
|
|
Title:
|
A Parallel Algorithm for Penalized Variable Selection in Gene-Environment Interactions
|
|
Author(s):
|
Yinhao Du* and Jie Ren and Fei Zhou and Cen Wu
|
|
Companies:
|
and Kansas State University and Kansas State University and Kansas State University
|
|
Keywords:
|
penalized variable selection;
gene-environment interaction;
distributed ADMM;
high dimensional data
|
|
Abstract:
|
Gene-environment (G×E) interaction is critical for understanding the genetic basis of complex disease beyond genetic and environment main effects. Existing penalized variable selection approaches, despite success, still have limitations in efficiently handling large scale data. In this study, we propose a distributed alternating direction method of multipliers (ADMM) algorithm to achieve efficient and fast computation for identifying G×E interaction in penalized regression framework. Simulation studies show the proposed method has satisfactory identification performance and reduces more computation time than existing approaches. Further, the proposed method is adopted to analyze high dimensional genomics data from a case-control study. The identification of G×E interactions lead to biologically sensible findings and improved prediction.
|
Authors who are presenting talks have a * after their name.
