|
Activity Number:
|
23
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #307415 |
|
Title:
|
Probe-Level Modeling and Multiple Testing of Microarray Gene Expression
|
|
Author(s):
|
Tao Wang*+ and Magali Mouy and Jason Hsu and Hakon Hakonarson and Kari Stefansson
|
|
Companies:
|
University of South Florida and deCODE genetics and The Ohio State University and deCODE genetics and deCODE genetics
|
|
Address:
|
13201 Bruce B Downs Blvd., Tampa, FL, 33612,
|
|
Keywords:
|
microarray ; gene expression ; modeling ; multiple testing
|
|
Abstract:
|
In this talk, I present a novel statistical approach for oligonucleotide microarray data analysis that models the probe level data. Comparisons of our probe level analysis with the popular gene level analysis using R package affy and multtest found our approach more powerful in both detecting differences and establishing equivalence in gene expressions. This talk also addressed some important issues in analyzing gene expression data such as the handling for non-normality and unequal variances of the error distribution in the linear model. An real microarray experiment to study the potential effect of an additional cycle of amplification on gene expression signals were analyzed using both probe and gene level analysis for comparison.
|
