Global modeling of a bacterial metabolism

The experimental work undertaken at Genoscope in the context of the Metabolic Thesaurus project provides a particularly favorable context for the global modeling of the metabolism of the bacterium Acinetobacter baylyi. The reconstruction of a first version of the model from genomic annotation necessitated the development of tools for the extraction of metabolic data from the BioCyc database [* M. Krummenacker & al. (2005)] as well as a major curation effort. The accuracy of this model was evaluated qualitatively by comparison with an ensemble of phenotypic data from mutants, and quantitatively, by a series of sustained growth experiments in A. baylyi. The differences identified between predictions and experimental data have been systematically exploited to refine the model. Two cycles of reconstruction/coherence verification have produced the present state which consists of 638 reactions, implicating 563 metabolites, and 710 genes classified into 7 functional categories. This model has the goal of helping elucidate the metabolism of A. baylyi, as well as the interpretation of new experimental data generated in the context of the Metabolic Thesaurus project. Its completeness, precision and predictive capacities continue to evolve in rhythm with the project.
 
 

Reconstructed metabolic network of Acinetobacter baylyi.
(Click on the image to enlarge)

 
  Bibliography :
[*] M. Krummenacker, S. Paley, L. Mueller, T. Yan, and P.D. Karp.
  “Querying and Computing with BioCyc Databases.”
  Bioinformatics. 2005. 21:3454-5.