Linear Coherent Bi-Clustering via Beam Searching and Sample Set Clustering

We propose a new bi-clustering algorithm, LinCoh, for finding linear coherent bi-clusters in gene expression microarray data. Our method exploits a robust technique for identifying conditionally correlated genes, combined with an efficient density based search for clustering sample sets. Experimental results on both synthetic and real datasets demonstrated that LinCoh consistently finds more accurate and higher quality bi-clusters than existing bi-clustering algorithms.