Department of Molecular Genetics and Microbiology, University of New Mexico Health Sciences Center, Albuquerque, NM 87131, USA. †Department of Biology, University of New Mexico, Albuquerque, NM 87131, USA. ‡Current address: DOE Joint Genome Institute, 2800 Mitchell Drive, Bldg 400, Walnut Creek, CA 94596, USA. Correspondence: Stephanie W Ruby. E-mail: sruby@ reviews Published: 30 March 2004 Genome Biology 2004, 5:R29 The electronic version of this article is the complete one and can be found online at Received: 27 November 2003 Revised: 29 January 2004 Accepted: 12 February 2004. | Method Open Access Enriching for direct regulatory targets in perturbed gene-expression profiles Susannah G Tringe Andreas Wagner and Stephanie W Ruby Addresses Department of Molecular Genetics and Microbiology University of New Mexico Health Sciences Center Albuquerque NM 87131 USA. ỶDepartment of Biology University of New Mexico Albuquerque NM 87131 USA. Current address DOE Joint Genome Institute 2800 Mitchell Drive Bldg 400 Walnut Creek CA 94596 USA. Correspondence StephanieWRuby. E-mail sruby@ Published 30 March 2004 Genome Biology 2004 5 R29 The electronic version of this article is the complete one and can be found online at http 2004 5M R29 Received 27 November 2003 Revised 29 January 2004 Accepted 12 February 2004 2004 Tringe et al. licensee BioMed Central Ltd. This is an Open Access article verbatim copying and redistribution of this article are permitted in all media for any purpose provided this notice is preserved along with the article s original URL. Abstract Here we build on a previously proposed algorithm to infer direct regulatory relationships using gene-expression profiles from cells in which individual genes are deleted or overexpressed. The updated algorithm can process networks containing feedback loops incorporate positive and negative regulatory relationships during network reconstruction and utilize data from double mutants to resolve ambiguous regulatory relationships. When applied to experimental data the reconstruction procedure preferentially retains direct transcription factor-target relationships. Background Gene-expression studies using cDNA or oligonucleotide arrays hold promise for elucidating the structure of genetic regulatory networks. A wealth of computational techniques have been proposed for extracting regulatory relationships from these data many of which rely on correlated expression patterns to identify temporally co-regulated genes reviewed in 1 2 . While these methods often detect important .
