Gene clustering analysis of DNA microarray data

Abstract

Although clustering analysis is by no means a new subject in the statistical literature, the large and complex multivariate datasets generated by microarray experiments have raised new methodological and computational challenges. This paper reviews several kinds of clustering methods and presents their merits and defects in regard to gene expression data from DNA microarray experiments, including supervised clustering methods, unsupervised clustering methods and model-based clustering methods. It is urgent to develop more feasible and realistic clustering methods specific for the analysis of the microarray experiment data.