Enriching regulatory networks by bootstrap learning using optimised GO-based gene similarity and gene links mined from PubMed abstracts

Ronald C. Taylor, Antonio Sanfilippo, Jason E. McDermott, Bob Baddeley, Roderick Riensche, Russell Jensen, Marc Verhagen, James Pustejovsky

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Increasingly, reverse engineering methods have been employed to infer transcriptional regulatory networks from gene expression data. Enrichment with independent evidence from sources such as the biomedical literature and the Gene Ontology (GO) is desirable to corroborate, annotate and expand these networks as well as manually constructed networks. In this paper, we explore a novel approach for computer-assisted enrichment of regulatory networks. GO-based gene similarity is first tuned to an initial network augmented with gene links mined from PubMed and then used to drive network construction using a bootstrapping algorithm. We describe two applications of this approach and discuss its added value in terms of corroboration, annotation and expansion of manually constructed and reversed engineered networks.

Original languageEnglish
Pages (from-to)56-82
Number of pages27
JournalInternational Journal of Computational Biology and Drug Design
Volume4
Issue number1
DOIs
Publication statusPublished - Feb 2011
Externally publishedYes

    Fingerprint

Keywords

  • Biological network inference
  • Bootstrap learning
  • Computational biology
  • Gene expression analysis
  • Gene Ontology
  • Network enrichment
  • PubMed
  • Regulatory networks
  • Simulated annealing

ASJC Scopus subject areas

  • Computer Science Applications
  • Drug Discovery

Cite this