|
Activity Number:
|
609
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #305186 |
|
Title:
|
A Semiparametric Unified Approach for the Detection of Differential Gene Expression in Microarrays
|
|
Author(s):
|
Jan R. De Neve*+ and Olivier Thas and Lieven Clement and Jean-Pierre Ottoy
|
|
Companies:
|
Ghent University and Ghent University and Ghent University and Ghent University
|
|
Address:
|
Coupure Links 653, Ghent, International, 9000, Belgium
|
|
Keywords:
|
microarrays ; stochastic ordering ; semi-parametric inference ; differential gene expression
|
|
Abstract:
|
A general method is proposed for detecting differential genes in high density oligonucleotide microarrays. It is a unified approach in the sense that it integrates the three preprocessing steps and the statistical testing methods into one semiparametric model. An important characteristic is that no stringent assumptions are imposed on the background correction and normalization steps. Instead of focusing on mean differences in gene expression, we formulate the model in terms of stochastic ordering. In particular, probabilities $P(Y_1 < Y_2 )$, with $Y_i$ the intensity of a gene in group $i$ ($i = 1, 2$), are modeled in terms of predictor variables. We present some theoretical results and spike-in studies are considered for comparing the performance of this new method with existing methods. Finally we apply the new method to a publicly available data set.
|
