Statistical Analysis for Detecting Differentially Expressed Genes

A common task in analyzing microarray data is to determine which genes are differentially expressed across two tissue samples, such as normal and tumor tissues, or samples obtained under two experimental conditions. In this paper, we point out a problem with previous method(Zhao and Pan, 2003) of constructing a null and test statistics. We propose resampling approach to construct test statistic as an alternative method. We also present results of simulation data and real data analysis.