Coclustering Under Gene Ontology Derived Constraints for Pathway Identification

In this chapter the authors propose a new methodology that minimizes the intervention of the analyst within the coclustering process and that provides meaningful coclusters whose discovery and interpretation are enhanced by embedding gene ontology (GO) annotations. To show the effectiveness of this approach, the authors apply their methodology on a gene expression data set consisting on different stress conditions for Saccharomyces cerevisiae (baker's yeast). The chapter explores the related bibliography in details. Many coclustering methods have been developed, possibly dedicated to gene expression data analysis. Next, the chapter briefly introduces the constrained coclustering algorithm the authors use in their framework. Then, the chapter describes the methodology for coclustering using GO-derived constraints. Finally, the chapter presents a case study on a real gene expression data set.