A Two-Step Hierarchical Hypothesis Set Testing Framework, with Applications to Gene Expression Data on Ordered Categories

• Activity Number: 642
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Yihan Li*+ and Debashis Ghosh
• Companies: Pennsylvania State University and Penn State University
• Keywords: Multiple testing ; Overall false discovery rate ; Microarray ; Time-course ; Dose-response

Abstract

In complex large-scale experiments, in addition to simultaneously considering a large number of features, multiple hypotheses are often being tested for each feature. This leads to the problem of multi-dimensional multiple testing. In this paper, we consider a framework for testing multiple sets of hypothesis. We adopt the concept of the overall false discovery rate (OFDR) by Benjamini and Heller (2008) for controlling false discoveries on the hypothesis set level. Extending a procedure by Heller et al. (2009), we propose a general two-step hierarchical hypothesis set testing procedure, which controls the OFDR under independence across hypothesis sets. We applied the framework to microarray time-course/dose-response experiments, and proposed three procedures for testing differential expression and making multiple directional decisions for each gene. Simulation studies confirm our procedures' control of the OFDR and show that two of our new procedures achieve higher power than previous methods. Finally, the proposed methodology is applied to a microarray dose-response study by Coser et al. (2003) for identifying estrogen sensitive genes that are induced at low concentrations of E2.