Validating Gene Clusterings by Selecting Informative Gene Ontology Terms with Mutual Information
I.G. Costa, M.C.P. de Souto and A. Schliep
In Advances in Bioinformatics and Computational Biology, Proceedings of the Brazilian Symposium on Bioinformatics, Springer Verlag, 81–92, 2007.
We propose a method for global validation of gene clusterings. The method selects a set of informative and non-redundant GO terms through an exploration of the Gene Ontology structure guided by mutual information. Our approach yields a global assessment of the clustering quality, and a higher level interpretation for the clusters, as it relates GO terms with specific clusters. We show that in two gene expression data sets our method offers an improvement over previous approaches.
A reprint is available as PDF.
The publication includes results from the following projects or software tools: MASCAAT.
The following presentation(s) are based on this publication: Aug. 29, 2007 by Ivan Costa at Brazilian Symposium on Bioinformatics (Contributed Talk).