Title: Integration of clustering and biclustering procedures for analysis of large DNA microarray datasets
Authors: Doris Z. Wang, Hong Yan
Addresses: Department of Electronic Engineering, City University of Hong Kong, Kowloon, Hong Kong. ' Department of Electronic Engineering, City University of Hong Kong, Kowloon, Hong Kong
Abstract: In this paper, a new biclustering method is introduced based on a geometric method. The Hough Transform (HT) in column-pair space is used to find sub-biclusters and the k-means method is used to optimise the combining process. The experiment results show that the K-means Based Geometric Biclustering (KGBC) algorithm proposed here reduces the computing time for combining sub-biclusters and provides considerably higher classification accuracy compared with existing methods. Experiments on both simulated and real microarray data demonstrate that our method can identify biclusters with different noise levels and overlapping degrees.
Keywords: microarray data analysis; biclustering; Hough transform; k-means clustering; DNA microarray datasets; classification accuracy.
International Journal of Computational Biology and Drug Design, 2011 Vol.4 No.2, pp.179 - 193
Published online: 24 Jan 2015 *Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article