|
Activity Number:
|
76
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 8:00 PM to 9:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #307366 |
|
Title:
|
Multivariate Simulation of Gene Expression Data
|
|
Author(s):
|
Rudolph Parrish*+ and Horace J. Spencer
|
|
Companies:
|
University of Louisville and University of Arkansas for Medical Sciences
|
|
Address:
|
School of Public Health, Louisville, KY, 40292,
|
|
Keywords:
|
microarrays ; distribution modeling ; simulation ; multivariate ; normalization
|
|
Abstract:
|
Simulation of microarray data in a realistic form is valuable for the purpose of evaluating the many different proposed techniques for pre-processing and analyzing such data and for designing experiments. For example, normalization procedures can be investigated to determine their relative statistical properties. Mathematical and statistical distribution modeling methods can be applied advantageously to help address this problem. A computational method is described in which variation among arrays can be characterized simultaneously for a large number of genes resulting in a multivariate model of gene expression which incorporates correlations. This model is used to simulate microarray expression data and probe intensity data and then to compare different normalization approaches.
|
