A Comparison of Methods for Identifying Differentially Expressed Genes in Microarray Experiments

• Activity Number: 404
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Manel Wijesinha*+ and Dhammika Amaratunga
• Companies: Penn State University and Johnson & Johnson Pharmaceutical R&D, LLC
• Keywords: conditional t ; limma ; microarray

Abstract

A common task in microarray experiments is to identify genes that are differentially expressed across two conditions, e.g., in normal tissue vs diseased tissue. This can be addressed quite simply by doing a series of ordinary t-tests. However, several authors have argued that the power of the tests can be improved by borrowing strength across the genes, a process that leads to modified t-tests. Strength can be borrowed either parametrically, leading to methods such as limma, or semi-parametrically, leading to methods such as Conditional t. In this work, we will compare the effectiveness of these methods in situations where standard assumptions, such as normality, hold and in situations where such assumptions do not hold. In addition, we will explore some variations of the volcano plot for gene selection.