Reassessing design and analysis of two-colour microarray
experiments using mixed effects models.
Departments of Animal Science and Fisheries & Wildlife
Michigan State University
Friday, March 11, 2005, 12:00 - 1:00 pm
Gene expression microarray studies have lead to interesting experimental design and statistical analysis challenges. The comparison of expression profiles across groups or populations (within experimental or observational settings) is one of the most common objectives of microarrays experiments. In this presentation we will discuss some issues regarding design and statistical analysis for two color microarray platforms using mixed linear models, with special attention directed towards a distinction of the different hierarchical levels of replication and the consequent effect on the use of appropriate error terms for comparing experimental groups. We will review the traditional analysis of variance (ANOVA) models proposed for microarray data with their extensions to hierarchically replicated experiments. We will also suggest and compare linear mixed models suitable for the analysis of either log ratios or log intensityvalues, denoting the similarity between both approaches in the presence of multiple sources of variability. In addition, we will discuss the relative efficiency of different experimental designs for microarray experiments within a hierarchical replication context.