Biclustering of gene expression data

Biclustering is an important problem that arises in diverse applications, including the analysis of gene expression and drug interaction data. A large number of clustering approaches have been proposed for gene expression data obtained from microarray experiments. However, the results from the application of standard clustering methods to genes are limited. This limitation is imposed by the existence of a number of experimental conditions or gene samples, where the expression levels of the same genes are uncorrelated. A similar limitation exists when conditionclustering is performed. The goal of biclustering is to find submatrices of genes and conditions, or samples where the genes have nearly the same expression levels for nearly all conditions. Some clustering methods have been adopted or proposed. However, some concerns still remain, such as the robustness of mining methods on the noise and input parameters. In this paper we tackle the problem of effectively clustering gene expression data by proposing an algorithm. We use a density-based approach to identify clusters. Our experimental results show that the algorithm is effective. Key-Words: -data mining, clustering, biclustering