An Empirical Bayesian Method for Oligonucleotide Microarray Data

• Activity Number: 288
• Type: Contributed
• Date/Time: Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
• Sponsor: ENAR
• Author(s): Yan Lin*+ and Eleanor Feingold
• Companies: University of Pittsburgh and University of Pittsburgh

Abstract

Gene expression microarrays have become powerful tools in many areas of biological and biomedical research. One of the most common uses of microarrays is simply an exploratory study to compare two treatments (e.g., tumor and normal tissue) and look for a list of genes that might be differentially expressed between the two. Differential expression is typically measured by computing a t-statistic or similar statistics for each gene. The genes are then ranked according to the absolute value of the t-statistic, and the 20 or 50 best candidates might be studied in follow-up experiments. When sample sizes are small, the t-statistic can be problematic, because variances are estimated poorly and the "top 20" list is often dominated by the genes with the lowest variance estimates. Lonnstedt and Speed (2001) proposed an empirical Bayesian method for avoiding this problem. However, their method was designed mainly for the two-colored cDNA microarray, which produces paired data. We propose a simplification for Lonnstedt and Speed's method, and then extend it to unpaired data with two or more independent treatments. We demonstrate our method on both simulated and real data.