Activity Number:
|
192
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, August 13, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
ENAR
|
Abstract - #300119 |
Title:
|
Detecting Differentially Expressed Genes with Microarrays Using Bayesian Model Selection
|
Author(s):
|
J. Rao*+ and Hemant Ishwaran
|
Affiliation(s):
|
Case Western Reserve University and Cleveland Clinic Foundation
|
Address:
|
10900 Euclid Avenue, Cleveland, Ohio, 44106, USA
|
Keywords:
|
oligonuceotide microarrays ; genetic signature
|
Abstract:
|
Oligonucleotide microarrays are amongst a set of technologies that allow for high throughput assessment of vast numbers of gene expressions simultaneously. Detection of differentially expressed genes between two experimental groups is a typical question of importance to scientists. This information provides evidence of genetic signatures that may be involved in a disease process. Usual forms of analysis include empirical Bayes analyses and multiple hypothesis testing using new approaches to control false detection rates (FDR) rather than family-wise error rates. In this paper, we provide an alternative strategy using high dimensional Bayesian model selection based on an adaptation of work first shown in Ishwaran and Rao (2000). The Bayesian model selection process allows for model averaging, which pays off especially for moderate expressing genes working to better separate those genes that are differentially expressing from those that are not. The methods will be illustrated on simulated data as well as a large repository of array data collected on metastatic and normal colon cancer samples at CWRU.
|