Title: Recent Patents on Biclustering Algorithms for Gene Expression Data Analysis
Volume: 5
Issue: 2
Author(s): Alan Wee-Chung Liew, Ngai-Fong Law and Hong Yan
Affiliation:
Keywords:
Biclustering, cluster analysis, multidimensional data analysis, gene expression data, geometric-based biclustering, microarray data, pattern discovery, hierarchical clustering, localized groupings, significant homogeneity, Statistical-Algorithmic Method for Bicluster Analysis, biclustering algorithms, multiplicative coherent values, data matrix, Gaussian mixtures
Abstract: In DNA microarray experiments, discovering groups of genes that share similar transcriptional characteristics is instrumental in functional annotation, tissue classification and motif identification. However, in many situations a subset of genes only exhibits a consistent pattern over a subset of conditions. Although used extensively in gene expression data analysis, conventional clustering algorithms that consider the entire row or column in an expression matrix can therefore fail to detect useful patterns in the data. Recently, biclustering has been proposed as a powerful computational tool to detect subsets of genes that exhibit consistent pattern over subsets of conditions. In this article, we review several recent patents in bicluster analysis, and in particular, highlight a recent patent from our group about a novel geometric-based biclustering method that handles the class of bicluster patterns with linear coherent variation across the row and/or column dimension. This class of bicluster patterns is of particular importance since it subsumes all constant, additive, and multiplicative bicluster patterns normally used in gene expression data analysis.
