|
Activity Number:
|
445
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
International Indian Statistical Association
|
| Abstract - #303193 |
|
Title:
|
Multiple Testing Procedures for Integrative Genomics Problems in Cancer Studies
|
|
Author(s):
|
Debashis Ghosh*+ and Laila M. Poisson
|
|
Companies:
|
Penn State University and University of Michigan
|
|
Address:
|
514A Wartik Lab, University Park, PA, 16802,
|
|
Keywords:
|
Biomarkers ; False discovery rate ; Outlier profile analysis ; Multiple comparisons ; Multivariate analysis ; Oncology
|
|
Abstract:
|
With the explosion of new molecular technologies available for assaying different biological activities of cancer samples, statisticians and data analysts are being faced with multiple layers of biological data. In this talk, we focus on the issue of integrating data sets in which multiple molecular activities have been assessed on the same set of samples. For most of these studies the goal is to prioritize molecules and/or pathways for further experimentation. In such a situation, it is natural to use multiple testing procedures. We consider several issues in the use of multiple testing procedures with data from several platforms from matched clinical samples with attendant information. These include pathway-directed analyses, outlier detection methods, and multivariate multiple testing procedures. Real data examples will be used to illustrate the proposed methodologies.
|
