Mib: Using Mutual Information for Biclustering High Dimensional Data

Most of the biclustering algorithms for gene expression data are based either on the Euclidean distance or correlation coefficient which capture only linear relationships. However, in gene expression data, non linear relationships may exist between the genes. Mutual Information between two variables provides a more general criterion to investigate dependencies amongst variables. In this paper, we propose an algorithm that uses mutual information for biclustering gene expression data. We present the experimental results on synthetic data. None of the distance based biclustering algorithms will identify the biclusters in our synthetic data which our algorithm is able to report. In future we intend to use our algorithm on gene expression data. KEYWORDS Biclustering and Mutual Information.

• Type: Conference
• Year: 2008
• Where: IADIS
• Authors: Neelima Gupta, Seema Aggarwal
• Keywords: Algorithm, Gene Expression Data, IADIS 2008, Internet Technology, Mutual Information