Clustering For Studying DNA Microarray Data

DNA microarrays are used to produce large sets of expression measurements from which specific biological information is sought. And efficient and reliable algorithms for clustering are required when analyzing microarray data. Most of the conventional methods group genes or samples with similar expression profiles into the same cluster. Due to the characteristics of microarray data that it has a few tissue samples but each corresponds to expression levels of a huge number of genes, clustering algorithms based on DNA microarray is performed in a space of superhigh dimensions, resulting in the problem of curse of dimensionality. Focusing on relationship between tissues or genes, a novel clustering method is proposed in this paper, which combines the classical correlation-based clustering for studying DNA microarray data and graphical theory. It can penetrate into the similarity between tissues or genes, while avoiding the problem of curse of dimensionality. Application of the proposed method to publicly available yeast and NCI data, compared with some other methods, demonstrates that the proposed method is more effective and efficient in clustering genes or tissues.