|
Activity Number:
|
283
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 14, 2002 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Stat. Sciences*
|
| Abstract - #301795 |
|
Title:
|
Hierarchical Bayes De-Noising of Microarray Experiments
|
|
Author(s):
|
Stephen Erickson*+ and Chiara Sabatti+
|
|
Affiliation(s):
|
University of California, Los Angeles and University of California, Los Angeles
|
|
Address:
|
8130 Math Sciences, Los Angeles, California, 90095-1554, USA 695 Charles Young Drive South, Los Angeles, California, 90095-7088, USA
|
|
Keywords:
|
False Discovery Rate ; Multiple Comparison ; Thresholding ; Shrinkage
|
|
Abstract:
|
We present a cDNA microarray experiment in which the gene expressions of mouse neurospheres cultured with two different growth factors are compared. Because of the highly stochastic nature of the experiments, it is difficult to distinguish between signal and noise in assessing the variation of expression of the genes between the two cell lines.
Often a hypothesis testing point of view is taken, where multiple comparison procedures are adopted to control either the Family-wise Error Rate or the False Discovery Rate. Rather than answering the question of which genes change expression in a testing framework, however, we want to take a de-noising approach that may lead to an estimate of the true change in gene expression under the two conditions. We use a hierarchical Bayes framework to account for slide and experiment variability. We choose our priors on the true change in expression to represent the fact that a large portion of the genes is not expected to undergo any change in expression when the cells are grown under the two different conditions. The prior assumptions provide a link between our estimation procedure and a hypothesis testing one.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |
