Interface 2003
Abstract

Exploratory Detection of Differential Gene Expression
John D. Storey, (University of California, Berkeley), storey@stat.berkeley.edu

Abstract

We propose a statistical method for detecting differentially expressed genes in DNA microarray experiments that draws on ideas from hypothesis testing, machine learning, and false discovery rates. The final product of this methodology is a listing of genes in order of evidence for differential gene expression, as well as a gene-wise measure of significance called the q-value. Some simple arguments indicate that this method is the most powerful among all distribution free approaches to detecting differential gene expression. Numerical evidence indicates that this method outperforms the most widely used methods.


Take me back to the main conference page.