Method

Application of independent component analysis to microarrays

Su-In Lee¹ and Serafim Batzoglou²

¹  Department of Electrical Engineering, Stanford University, Stanford, CA94305-9010, USA

²  Department of Computer Science, Stanford University, Stanford, CA94305-9010, USA

author email corresponding author email

Genome Biology 2003, 4:R76doi:10.1186/gb-2003-4-11-r76

Published:	24 October 2003

Subject areas: Bioinformatics, Methods, Genome studies, Model organisms

Abstract

We apply linear and nonlinear independent component analysis (ICA) to project microarray data into statistically independent components that correspond to putative biological processes, and to cluster genes according to over- or under-expression in each component. We test the statistical significance of enrichment of gene annotations within clusters. ICA outperforms other leading methods, such as principal component analysis, k-means clustering and the Plaid model, in constructing functionally coherent clusters on microarray datasets from Saccharomyces cerevisiae, Caenorhabditis elegans and human.