Biclustering Algorithms for Microarray Data

Abstract or Introduction

This survey aims to analyze a large number of existing methods to biclustering, and classify them in accordance with the methods used to perform the search.

DNA microarray aims at extracting useful information that can be applied in medical and biological studies. Clustering is one of the most utilized data mining technique to analyze the gene expression data. Generally, there are subsets of genes that behave similarily or under subsets of conditions. Therefore, biclustering is introduced to identify the subgroups of genes and subgroups of conditions by performing simultaneous clustering of both rows and columns of the gene expression matrix, rather than clustering these two dimensions separately.

DNA microarray technologies have made it feasible to monitor transcription levels of tens of thousands of genes in a single expriement. A typical DNA microarray experiment involves a multistep procedure: fabrication of microarrays by fixing properly designed oligonucleotides representing specific genes; hybridization of cDNA populations onto the microarray; scanning hybridization signals and image analysis; transformation and normalization of data; and analyzing data to identify differentially expressed genes as well as sets of genes that are co-regulated.