Extraction of Optimal Biclusters from Gene Expression Data

Abstract

Biclustering is a data mining technique which allows simultaneous clustering of the rows and columns of a matrix. In this paper, MAXimal BICluster algorithm (MAXBIC) identifies coherent biclusters of maximum size with high Average Spearman Rho (ASR). This proposed query based algorithm includes three steps viz. three tier pre-processing, identifying a bicluster seed and growing the seed till an optimal bicluster is obtained. Experimental results show the effectiveness of the proposed algorithm. © Springer-Verlag Berlin Heidelberg 2010.