Bioinformatics Advance Access originally published online on March 25, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Bioinformatics 20(12) © Oxford University Press 2004; all rights reserved.
Hypothesis-driven approach to predict transcriptional units from gene expression data
Max Planck Institute of Molecular Plant Physiology, 14476 Golm, Germany
Received on August 25, 2003; revised on February 24, 2004; accepted on February 25, 2004
Advance Access Publication March 25, 2004
Motivation: A major issue in computational biology is the reconstruction of functional relationships among genes, for example the definition of regulatory or biochemical pathways. One step towards this aim is the elucidation of transcriptional units, which are characterized by co-responding changes in mRNA expression levels. These units of genes will allow the generation of hypotheses about respective functional interrelationships. Thus, the focus of analysis currently moves from well-established functional assignment through comparison of protein and DNA sequences towards analysis of transcriptional co-response. Tools that allow deducing common control of gene expression have the potential to complement and extend routine BLAST comparisons, because gene function may be inferred from common transcriptional control.
Results: We present a co-clustering strategy of genome sequence information and gene expression data, which was applied to identify transcriptional units within diverse compendia of expression profiles. The phenomenon of prokaryotic operons was selected as an ideal test case to generate well-founded hypotheses about transcriptional units. The existence of overlapping and ambiguous operon definitions allowed the investigation of constitutive and conditional expression of transcriptional units in independent gene expression experiments of Escherichia coli. Our approach allowed identification of operons with high accuracy. Furthermore, both constitutive mRNA co-response as well as conditional differences became apparent. Thus, we were able to generate insight into the possible biological relevance of gene co-response. We conclude that the suggested strategy will be amenable in general to the identification of transcriptional units beyond the chosen example of E.coli operons.
Availability: The analyses of E.coli transcript data presented here are available upon request or at http://csbdb.mpimp-golm.mpg.de/
Contact: Steinhauser{at}mpimp-golm.mpg.de
* To whom correspondence should be addressed.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. W. W. Brouwer, O. P. Kuipers, and S. A. F. T. van Hijum The relative value of operon predictions Brief Bioinform, September 1, 2008; 9(5): 367 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Horan, C. Jang, J. Bailey-Serres, R. Mittler, C. Shelton, J. F. Harper, J.-K. Zhu, J. C. Cushman, M. Gollery, and T. Girke Annotating Genes of Known and Unknown Function by Large-Scale Coexpression Analysis Plant Physiology, May 1, 2008; 147(1): 41 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Voll, A. Jamai, P. Renne, H. Voll, C. R. McClung, and A. P.M. Weber The Photorespiratory Arabidopsis shm1 Mutant Is Deficient in SHM1 Plant Physiology, January 1, 2006; 140(1): 59 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lisso, D. Steinhauser, T. Altmann, J. Kopka, and C. Mussig Identification of brassinosteroid-related genes by means of transcript co-response analyses Nucleic Acids Res., May 12, 2005; 33(8): 2685 - 2696. [Abstract] [Full Text] [PDF]
|
|