Discriminative local subspaces in gene expression data for effective gene function prediction. http://dx.doi.org/10.1093/bioinformatics/bts455

Abstract

Motivation: Massive amounts of genome-wide gene expression data have become available, motivating the development of computational approaches that leverage this information to predict gene function. Among successful approaches, supervised machine learning methods, such as Support Vector Machines, have shown superior prediction accuracy. However, these methods lack the simple biological intuition provided by coexpression networks, limiting their practical usefulness.

Results: In this work we present Discriminative Local Subspaces (DLS), a novel method that combines supervised machine learning and coexpression techniques with the goal of systematically predict genes involved in specific biological processes of interest. Unlike traditional coexpression networks, DLS uses the knowledge available in Gene Ontology (GO) to generate informative training sets that guide the discovery of expression signatures: expression patterns that are discriminative for genes involved in the biological process of interest. By linking genes coexpressed with these signatures, DLS is able to construct a discriminative coexpression network that links both, known and previously uncharacterized genes, for the selected biological process. This paper focuses on the algorithm behind DLS and shows its predictive power using an Arabidopsis thaliana dataset and a representative set of 101 GO-terms from the Biological Process Ontology. Our results show that DLS has a superior average accuracy than both, Support Vector Machines and Coexpression Networks. Thus, DLS is able to provide the prediction accuracy of supervised learning methods, while maintaining the intuitive understanding of coexpression networks.