|
Activity Number:
|
380
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303375 |
|
Title:
|
Statistical and Numerical Dependence in Gene Expression Summaries
|
|
Author(s):
|
John R. Stevens*+ and Gabriel Nicholas
|
|
Companies:
|
Utah State University and University of Wisconsin-Madison
|
|
Address:
|
3900 Old Main Hill, Logan, UT, 84322-3900,
|
|
Keywords:
|
microarray ; bootstrap ; dependence
|
|
Abstract:
|
Gene expression studies using microarray technology are now commonplace, and make use of a wide variety of statistical methods for data analysis. Statistical methods to test for differential expression traditionally assume the independence of a gene's expression estimates from multiple arrays. When certain preprocessing methods are used to obtain those estimates, this assumption can be violated. We apply the bootstrap to estimate the covariance of each gene's expression level estimates, and show that the resulting statistical dependence is biological in nature. We introduce a measure for numerical dependence for gene expression summaries and discuss the relative performance of several common preprocessing methods with respect to this measure.
|
