Title: "Comparison of the FDR, pFDR, and SAM for power and error rates for analysis of microarray experiments"
Speaker: Bill Muir, Department of Animal Science, Purdue University
Place: Smith Hall (SMTH) 108; April 12, 2005, Tuesday, 4:30pm

Abstract

The most common methods for finding differentially expressed genes in microarray experiments are the FDR, pFDR, and SAM. Programs to computing the FDR and pFDR using SAS will be presented. Pros and cons of procedures (simulations vs real data vs qRT=PCR) to evaluate these methods will be discussed. Using all these alternative evaluation procedures, the methods will be compared for power, type I and II error rates and robustness to assumptions.



Click here for a full scheule of BIOINFORMATICS SEMINARS, past and present.