A systematic comparison and evaluation of biclustering methods for gene expression data.

@article{citeulike:825683, abstract = {MOTIVATION: In recent years, there have been various efforts to overcome the limitations of standard clustering approaches for the analysis of gene expression data by grouping genes and samples simultaneously. The underlying concept, which is often referred to as biclustering, allows to identify sets of genes sharing compatible expression patterns across subsets of samples, and its usefulness has been demonstrated for different organisms and datasets. Several biclustering methods have been proposed in the literature; however, it is not clear how the different techniques compare with each other with respect to the biological relevance of the clusters as well as with other characteristics such as robustness and sensitivity to noise. Accordingly, no guidelines concerning the choice of the biclustering method are currently available. RESULTS: First, this paper provides a methodology for comparing and validating biclustering methods that includes a simple binary reference model. Although this model captures the essential features of most biclustering approaches, it is still simple enough to exactly determine all optimal groupings; to this end, we propose a fast divide-and-conquer algorithm (Bimax). Second, we evaluate the performance of five salient biclustering algorithms together with the reference model and a hierarchical clustering method on various synthetic and real datasets for Saccharomyces cerevisiae and Arabidopsis thaliana. The comparison reveals that (1) biclustering in general has advantages over a conventional hierarchical clustering approach, (2) there are considerable performance differences between the tested methods and (3) already the simple reference model delivers relevant patterns within all considered settings.}, address = {Computer Engineering and Networks Laboratory, ETH Zurich, 8092 Zurich, Switzerland.}, author = {Preli\'{c}, A. and Bleuler, S. and Zimmermann, P. and Wille, A. and B\"{u}hlmann, P. and Gruissem, W. and Hennig, L. and Thiele, L. and Zitzler, E. }, citeulike-article-id = {825683}, issn = {1367-4803}, journal = {Bioinformatics}, keywords = {biclustering, clustering, expression}, month = {May}, number = {9}, pages = {1122--1129}, posted-at = {2006-09-02 11:25:27}, priority = {2}, title = {A systematic comparison and evaluation of biclustering methods for gene expression data.}, url = {http://view.ncbi.nlm.nih.gov/pubmed/16500941}, volume = {22}, year = {2006} }

TY - JOUR ID - citeulike:825683 L3 - citeulike-article-id:825683 N2 - MOTIVATION: In recent years, there have been various efforts to overcome the limitations of standard clustering approaches for the analysis of gene expression data by grouping genes and samples simultaneously. The underlying concept, which is often referred to as biclustering, allows to identify sets of genes sharing compatible expression patterns across subsets of samples, and its usefulness has been demonstrated for different organisms and datasets. Several biclustering methods have been proposed in the literature; however, it is not clear how the different techniques compare with each other with respect to the biological relevance of the clusters as well as with other characteristics such as robustness and sensitivity to noise. Accordingly, no guidelines concerning the choice of the biclustering method are currently available. RESULTS: First, this paper provides a methodology for comparing and validating biclustering methods that includes a simple binary reference model. Although this model captures the essential features of most biclustering approaches, it is still simple enough to exactly determine all optimal groupings; to this end, we propose a fast divide-and-conquer algorithm (Bimax). Second, we evaluate the performance of five salient biclustering algorithms together with the reference model and a hierarchical clustering method on various synthetic and real datasets for Saccharomyces cerevisiae and Arabidopsis thaliana. The comparison reveals that (1) biclustering in general has advantages over a conventional hierarchical clustering approach, (2) there are considerable performance differences between the tested methods and (3) already the simple reference model delivers relevant patterns within all considered settings. IS - 9 JF - Bioinformatics SN - 1367-4803 AD - Computer Engineering and Networks Laboratory, ETH Zurich, 8092 Zurich, Switzerland. EP - 1129 TI - A systematic comparison and evaluation of biclustering methods for gene expression data. VL - 22 SP - 1122 KW - biclustering KW - clustering KW - expression AU - Prelić, A AU - Bleuler, S AU - Zimmermann, P AU - Wille, A AU - Bühlmann, P AU - Gruissem, W AU - Hennig, L AU - Thiele, L AU - Zitzler, E PY - 2006/05/01/ UR - http://view.ncbi.nlm.nih.gov/pubmed/16500941 ER -