Bioinformatics Advance Access originally published online on July 20, 2009
Bioinformatics 2009 25(20):2723-2729; doi:10.1093/bioinformatics/btp441
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
An optimization model for metabolic pathways
1CEIT and TECNUN, University of Navarra, Manuel de Lardizabal 15, 20018 San Sebastian, Spain and 2Mathematical Sciences, Brunel University, Uxbridge, UB8 3PH, UK
* To whom correspondence should be addressed.
| Abstract |
|---|
Motivation: Different mathematical methods have emerged in the post-genomic era to determine metabolic pathways. These methods can be divided into stoichiometric methods and path finding methods. In this paper we detail a novel optimization model, based upon integer linear programming, to determine metabolic pathways. Our model links reaction stoichiometry with path finding in a single approach. We test the ability of our model to determine 40 annotated Escherichia coli metabolic pathways. We show that our model is able to determine 36 of these 40 pathways in a computationally effective manner.
Contact: john.beasley{at}brunel.ac.uk
Supplementary information: Supplementary data are available at Bioinformatics online.
Received on December 15, 2008; revised on June 3, 2009; accepted on July 13, 2009