Biological cluster validity indices based on the gene ontology

Abstract

With the invention of biotechnological high throughput methods like DNA microarrays and the analysis of the resulting huge amounts of biological data, clustering algorithms gain new popularity. In practice the question arises, which clustering algorithm as well as which parameter set generates the most promising results. Little work is addressed to the question of evaluating and comparing the clustering results, especially according to their biological relevance, as well on distinguishing biologically interesting clusters from less interesting ones. This paper presents two cluster validity indices intended to evaluate clusterings of gene expression data in a biological manner. © Springer-Verlag Berlin Heidelberg 2005.