CSDL Home IEEE/ACM Transactions on Computational Biology and Bioinformatics 2009 vol.6 Issue No.04 - October-December

Issue No.04 - October-December (2009 vol.6)

pp: 594-604

Miquel Salicrú , Barcelona University, Spain

Sergi Vives , Barcelona University, Spain

Tian Zheng , Columbia University, New York

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TCBB.2008.106

ABSTRACT

Cluster analysis has proven to be a useful tool for investigating the association structure among genes in a microarray data set. There is a rich literature on cluster analysis and various techniques have been developed. Such analyses heavily depend on an appropriate (dis)similarity measure. In this paper, we introduce a general clustering approach based on the confidence interval inferential methodology, which is applied to gene expression data of microarray experiments. Emphasis is placed on data with low replication (three or five replicates). The proposed method makes more efficient use of the measured data and avoids the subjective choice of a dissimilarity measure. This new methodology, when applied to real data, provides an easy-to-use bioinformatics solution for the cluster analysis of microarray experiments with replicates (see the Appendix). Even though the method is presented under the framework of microarray experiments, it is a general algorithm that can be used to identify clusters in any situation. The method's performance is evaluated using simulated and publicly available data set. Our results also clearly show that our method is not an extension of the conventional clustering method based on correlation or euclidean distance.

INDEX TERMS

Clustering analysis, confidence interval, gene expression data.

CITATION

Miquel Salicrú, Sergi Vives, Tian Zheng, "Inferential Clustering Approach for Microarray Experiments with Replicated Measurements",

*IEEE/ACM Transactions on Computational Biology and Bioinformatics*, vol.6, no. 4, pp. 594-604, October-December 2009, doi:10.1109/TCBB.2008.106REFERENCES