|
Activity Number:
|
63
- Omics Data: Study Design, Power and Sample Size
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2018 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #329343
|
Presentation
|
|
Title:
|
Multivariate FDR Control for Omics Data Integration
|
|
Author(s):
|
Ali Shojaie* and Kasra Alishahi and Ahmad Reza Ehyaee
|
|
Companies:
|
University of Washington and Sharif University of Technology and Sharif University of Technology
|
|
Keywords:
|
False Discovery Rate;
Data Integration;
Omics;
Multivariate Hypothesis
|
|
Abstract:
|
Advances in high-throughput technologies have facilitated the collection of multiple types of omics measurements for the same individual. Integrating such multi-type omics data requires combining the evidence from diverse omics data types, in order to facilitate novel discoveries. To this end, we propose a new multivariate FDR controlling procedure that efficiently combines the outcome of hypothesis testing from multiple omics data types. Unlike existing FDR controlling approaches, the proposed approach offers non-asymptotic FDR control in multivariate settings. We present an optimal procedure that achieves minimum false negative rate among all procedures that control the FDR at the same level, and demonstrate its utility in omics data integration through simulated and real data examples.
|
Authors who are presenting talks have a * after their name.
