Identification of coherent patterns in gene expression data using an efficient biclustering algorithm and parallel coordinate visualization

Abstract

Background: The DNA microarray technology allows the measurement of expression levels of thousands of genes under tens/hundreds of different conditions. In microarray data, genes with similar functions usually co-express under certain conditions only [1]. Thus, biclustering which clusters genes and conditions simultaneously is preferred over the traditional clustering technique in discovering these coherent genes. Various biclustering algorithms have been developed using different bicluster formulations. Unfortunately, many useful formulations result in NP-complete problems. In this article, we investigate an efficient method for identifying a popular type of biclusters called additive model. Furthermore, parallel coordinate (PC) plots are used for bicluster visualization and analysis.

Iterative bicluster-based least square framework for estimation of missing values in microarray gene expression data

“…Often, the gene similarity is measured based on the similarity of the expression profiles across all experimental conditions [15]. In reality, genes are co-expressed under certain conditions only [16][17][18][19][20]. Hence the gene similarity should be measured by considering only those related experimental conditions, rather than all the conditions.…”

Iterated Local Search for Biclustering of Microarray Data

“…One of the most popular evaluation functions in the literature is called Mean Squared Residue (MSR) [10]. MSR has been used by several biclustering algorithms [9,13,23]. Yet MSR is known to be deficient to assess correctly the quality of certain types of biclusters like multiplicative models [1,25,29,9].…”

“…The systematic search approach includes greedy algorithms [6,9,10,29], divideand-conquer algorithms [17,26] and enumeration algorithms [4,20]. The metaheuristic approach includes neighbourhood-based algorithms [8], GRASP [12,13] and evolutionary algorithms [15,16,23].…”

Bagging for Biclustering: Application to Microarray Data

“…SAMBA, applied to a lymphoma dataset, produces biclusters representing new concrete biological associations. Cheng et al [5] have proposed the pCluster method that has the advantage it can identify both additive and multiplicative biclusters in presence of overlap. They validated their method on yeast cell-cycle dataset using Gene Ontology annotations.…”