|
Activity Number:
|
201
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #308844 |
|
Title:
|
Use of Satterthwaite Degree of Freedom in t-Statistic for Microarray Experiments with Fewer Replications
|
|
Author(s):
|
Yu Ding*+ and Damaraju Raghavarao
|
|
Companies:
|
Merck & Co., Inc. and Temple University
|
|
Address:
|
150 Oakland Avenue, Lansdale, PA, 19446,
|
|
Keywords:
|
Fewer Replications ; Microarray Experiments ; Satterthwaite Degree of Freedom
|
|
Abstract:
|
Many statistical methods to identify differentially expressed genes are based on a large number of arrays. However, many microarray experiments have fewer replications because of scarcely available resource. Most standard methods can not be used or are not accurate with small degrees of freedom for the tests. In this presentation, we use data from doubtfully expressed genes and unexpressed genes to construct a t-test, which has larger degree of freedom to identify differentially expressed genes for both direct and indirect design. We also apply our new methodology to ApoAI data. The results are promising.
|
