Extending knowledge of Escherichia coli metabolism by modeling and experiment.docVIP

Minireview Extending knowledge of Escherichia coli metabolism by modeling and experiment Eberhard O Voit* and Monica Riley? Addresses: *Medical University of South Carolina, Charleston, SC 29425, USA. ?Marine Biological Laboratory, Woods Hole, MA 02543, USA. Correspondence: Monica Riley. E-mail: mriley@ Published: 28 October 2003 Genome Biology 2003, 4:235 The electronic version of this article is the complete one and can be found online at /2003/4/11/235 ? 2003 BioMed Central Ltd Abstract One of the challenges for ‘post-genomic’ biology is the integration of data from many different sources. Two recent studies independently take steps towards this goal for Escherichia coli, using mathematical modeling and a combination of gene expression and protein levels to predict new gene functions and metabolic behaviors. It has become a platitude of the post-genomic era that a deluge of data is being produced and awaits both computa- tional/mathematical analysis and experimental verification. There is no reason to argue with this observation, yet it leads to two immediate follow-up questions. First, has the genomic era come to an end already? And second, what types of mathematical and computational models would be most beneficial for dealing with the rich streams of data? Two recent articles, by Reed et al. [1] and Corbin et al. [2], answer the first question in the negative: there is still a lot of genomic research to be done. These two articles show that, even for one of the best studied of organisms, Escherichia coli, there are still very many genes for which we know little beyond their sequence and location. We don’t know what their functions are, exactly which genes are actively functioning at any one time, and to what degree some might even be entirely dispensable. The experimental approach proposed by Corbin et al. [2] she