Quick Hierarchical Biclustering on Microarray Gene Expression Data

Mining biclusters that exhibit both consistent trends and trends with similar degrees of fluctuations is vital to bioinformatics research. However; existing biclustering methods are not very efficient and effective at mining such biclusters. Moreover, few inter-bicluster relationships are delivered to biologists. In this paper, we introduce a quick hierarchical biclustering algorithm (QHB) to efficiently mine biclusters with both consistent trends and trends with similar degrees of fluctuations. Our QHB produces not only biclusters but also a hierarchical graph of inter-bicluster relationships. We experimented with the Yeast dataset and compared QHB against an existing biclustering scheme, DBF Our results show that QHB identifies biclusters with better quality. In addition, QHB shows the relationships among biclusters. Moreover compared with DBF, QHB is much more efficient and offers users a progressive way of bicluster exploration