Activity Number:
|
465
- Biometrics and High-Dimensional Data
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biometrics Section
|
Abstract #324117
|
View Presentation
|
Title:
|
Score Test for Case-Cohort Design Applied to High-Throughput Gene Expression Analysis
|
Author(s):
|
Huining Kang* and John Carl Pesko
|
Companies:
|
University of New Mexico and University of New Mexico
|
Keywords:
|
Score test ;
Case-cohort design ;
High-throughput gene expression
|
Abstract:
|
The case-cohort (CCH) design is a cost-efficient method of analysis for time-to-failure data, particularly for large cohorts with low failure rates. This design consists of a randomly sampled subcohort augmented with all of the incident cases. The main advantage of this design is that the covariate data are only needed for the cases and subcohort controls rather than every subject in the full cohort. However, application of CCH design in high-throughput differential gene expression (DGE) analysis has not been seen in the public literature. We propose an analysis framework for DGE analysis under the CCH design using a score test for the proportional hazard model together with Benjamini and Hochberg (BH) procedure to adjust for multiple comparisons. The reason to use a score test is to avoid literately fitting a proportional hazard model for each gene. We also propose a stratified score test for DGE analysis that can be used to adjust for the effects of several categorical variables. The properties and performance of the proposed procedure are examined through application to a gene expression data set from a leukemia study and extensive simulation studies.
|
Authors who are presenting talks have a * after their name.