Biclustering of human cancermicroarraydata using co-similarity based co-clustering

 Biclustering of human cancermicroarraydata using co-similarity based co-clustering

Biclustering of gene expression data aims at finding localized patterns in a subspace. A bicluster (some-times called a co-cluster), in the context of gene expression data, is a set of genes that exhibit similar expression intensity under a subset of experimental features (conditions). Mostbiclustering algorithms proposed in the literature aim at findingsub-matricesthat exhibit some sort of coherence byselectingan initial sub-matrix and iteratively addingor subtracting rows and columns. These algorithms are generally dependent onthe initial, hardselection of the gene and condition clusters respectively. In this work, we adapt a recently proposed approach for clustering textual data to find biclusters in gene expression data. Our proposed technique is based onthe concept of co-similarity between genes (and between conditions) that exploits weightedhigher order paths in a bipartite graph representation of the gene expression data. Therefore, webuild statistical relations between genes and between conditions bycomparing all genes and conditions before finally extracting biclusters from the data. Weshow that the proposed technique is able to find meaningfulnon-overlapping biclusters bothonsynthetically generated data as w ell as realcancer data. Our results indicate that the proposed technique is resistant to noise in the data and can successfully retrieve biclusters even in the presenceof relatively large amount of noise