Gene Set Analysis with Covariates

In microarray experiments, expression profiles are obtained for thousands of genes under several treatments. Traditionally, most of the statistical techniques employed are concentrated around univariate methods. They ignore the inter-gene dependence and do not use any prior biological knowledge. Gene set analysis addresses both these concerns by analyzing together a group of correlated genes, for example genes that share a common biological function, chromosomal location, or regulation. In this paper we propose a multivariate analysis of covariance model (MANCOVA) for gene set analysis with covariates. Principal component analysis (PCA) is used to address the dimensionality problem. The two testing procedures presented are shown to perform well using simulations.