Design Challenges in Microarray Experiments

• Activity Number: 174
• Type: Invited
• Date/Time: Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Jason Hsu*+ and Patty Solomon
• Companies: Ohio State University and University of Adelaide
• Keywords: microarrays ; optimal experimental design ; gene expression

Abstract

Microarrays are powerful tools for surveying the expression levels of thousands of genes simultaneously and have found wide application in the biomedical, agricultural, and pharmaceutical sciences. There are, however, many sources of uncertainty in microarray experiments and good experimental design is essential for answering the questions of interest to biologists accurately and precisely. Unfortunately, the matter of choosing a suitable design cannot be addressed intuitively and although there is a large and well-established literature on the design of experiments, little is directly relevant to the microarray context. This presents a challenge to statisticians to find new methodology for generating and choosing designs. I will outline the key challenges and describe joint research with Gary Glonek on optimal designs for complex two-color microarray experiments. Our principles are to first identify the parameters of interest, then to seek designs that minimize the variance of the estimates subject to resource constraints. I will elucidate the issues with case studies on measuring gene expression in osteoarthritis, leukaemogenesis, and embryonic stem cells.