Analysis of the Effect of Reversibility Constraints on the Predictions of Genome-Scale Metabolic Models

  • José P. Faria
  • Miguel Rocha
  • Rick L. Stevens
  • Christopher S. Henry
Conference paper

DOI: 10.1007/978-3-642-13214-8_27

Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 74)
Cite this paper as:
Faria J.P., Rocha M., Stevens R.L., Henry C.S. (2010) Analysis of the Effect of Reversibility Constraints on the Predictions of Genome-Scale Metabolic Models. In: Rocha M.P., Riverola F.F., Shatkay H., Corchado J.M. (eds) Advances in Bioinformatics. Advances in Intelligent and Soft Computing, vol 74. Springer, Berlin, Heidelberg

Abstract

Reversibility constraints are one aspect of genome-scale metabolic models that has received significant attention recently. This study explores the impact of complete removal of reversibility constraints on the gene essentiality and growth phenotype predictions generated using three published genome-scale metabolic models: the iJR904, the iAF1260, and the iBsu1103. In all three models, the accuracy in predicting essential genes declined significantly with the relaxation of reversibility constraints, while the accuracy in predicting nonessential genes increased only for the iJR904 and iAF1260 model. Additionally, the number of inactive reactions in all models declined substantially with the relaxation of the reversibility constraints. This study rapidly reveals the extent to which the reversibility constraints included in a metabolic model have been optimized, and it indicates those incorrect model predictions that may be repaired and those correct model predictions that may be broken by increasing the number of reversible reactions in a model.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • José P. Faria
    • 1
    • 2
    • 3
  • Miguel Rocha
    • 1
  • Rick L. Stevens
    • 2
    • 3
  • Christopher S. Henry
    • 2
    • 3
  1. 1.Dep. Informatics/ CCTCUniversity of MinhoPortugal
  2. 2.Computation InstituteThe University of ChicagoChicagoUSA
  3. 3.Argonne National LaboratoryArgonneUSA

Personalised recommendations