Knowledge and Information Systems

, Volume 30, Issue 2, pp 341–358

BicFinder: a biclustering algorithm for microarray data analysis

Authors

    • UTIC, Higher School of Sciences and Technologies of TunisUniversity of Tunis
    • LERIAUniversity of Angers
  • Mourad Elloumi
    • UTIC, Higher School of Sciences and Technologies of TunisUniversity of Tunis
  • Jin-Kao Hao
    • LERIAUniversity of Angers
Regular Paper

DOI: 10.1007/s10115-011-0383-7

Cite this article as:
Ayadi, W., Elloumi, M. & Hao, J. Knowl Inf Syst (2012) 30: 341. doi:10.1007/s10115-011-0383-7

Abstract

In the context of microarray data analysis, biclustering allows the simultaneous identification of a maximum group of genes that show highly correlated expression patterns through a maximum group of experimental conditions (samples). This paper introduces a heuristic algorithm called BicFinder (The BicFinder software is available at: http://www.info.univ-angers.fr/pub/hao/BicFinder.html) for extracting biclusters from microarray data. BicFinder relies on a new evaluation function called Average Correspondence Similarity Index (ACSI) to assess the coherence of a given bicluster and utilizes a directed acyclic graph to construct its biclusters. The performance of BicFinder is evaluated on synthetic and three DNA microarray datasets. We test the biological significance using a gene annotation web-tool to show that our proposed algorithm is able to produce biologically relevant biclusters. Experimental results show that BicFinder is able to identify coherent and overlapping biclusters.

Keywords

Biclustering Heuristics Evaluation function Data mining Analysis of DNA microarray data

Copyright information

© Springer-Verlag London Limited 2011