On clustering of genes

Abstract

The availability of microarray technology at an affordable price makes it possible to determine expression of several thousand genes simultaneously. Gene expression can be clustered so as to infer the regulatory modules and functionality of a gene relative to one or more of the annotated genes of the same cluster. The outcome of clustering depends on the clustering method and the metric being used to measure the distance. In this paper we study the popular hierarchal clustering algorithm and verify how many of the genes in the same cluster share functionality. Further, we will also look into the supervised clustering method for satisfying hypotheses and view how many of these genes are functionally related. © Springen Verlag Berlin Heidelberg 2006.