Skip to content
Journal article

Noise-robust algorithm for identifying functionally associated biclusters from gene expression data

Ahn J, Yoon Y, Park S...(+3 more)

Information Sciences, vol. 181, issue 3 (2011) pp. 435-449

  • 6

    Readers

    Mendeley users who have this article in their library.
  • 9

    Citations

    Citations of this article.
  • 660

    Views

    ScienceDirect users who have downloaded this article.
Sign in to save reference

Abstract

Biclusters are subsets of genes that exhibit similar behavior over a set of conditions. A biclustering algorithm is a useful tool for uncovering groups of genes involved in the same cellular processes and groups of conditions under which these processes take place. In this paper, we propose a polynomial time algorithm to identify functionally highly correlated biclusters. Our algorithm identifies (1) gene sets that simultaneously exhibit additive, multiplicative, and combined patterns and allow high levels of noise, (2) multiple, possibly overlapped, and diverse gene sets, (3) biclusters that simultaneously exhibit negatively and positively correlated gene sets, and (4) gene sets for which the functional association is very high. We validate the level of functional association in our method by using the GO database, protein-protein interactions and KEGG pathways. ?? 2010 Elsevier Inc. All rights reserved.

Author-supplied keywords

  • Biclustering
  • Data mining
  • Gene expression data analysis
  • Knowledge discovery
  • Microarray analysis

Find this document

Get full text

Cite this document

Choose a citation style from the tabs below