Email updates

Keep up to date with the latest news and content from BMC Systems Biology and BioMed Central.

Open Access Highly Accessed Open Badges Research article

Analysis of complex metabolic behavior through pathway decomposition

Kuhn Ip12345, Caroline Colijn6 and Desmond S Lun12345*

Author Affiliations

1 Phenomics and Bioinformatics Research Centre, University of South Australia, Mawson Lakes, SA 5095, Australia

2 School of Mathematics and Statistics, University of South Australia, Mawson Lakes, SA 5095, Australia

3 Australian Centre for Plant Functional Genomics, University of South Australia, Mawson Lakes, SA 5095, Australia

4 Center for Computational and Integrative Biology, Rutgers University, Camden, NJ 08102, USA

5 Department of Computer Science, Rutgers University, Camden, NJ 08102, USA

6 Department of Engineering Mathematics, University of Bristol, Bristol BS8 1TR, UK

For all author emails, please log on.

BMC Systems Biology 2011, 5:91  doi:10.1186/1752-0509-5-91

Published: 3 June 2011



Understanding complex systems through decomposition into simple interacting components is a pervasive paradigm throughout modern science and engineering. For cellular metabolism, complexity can be reduced by decomposition into pathways with particular biochemical functions, and the concept of elementary flux modes provides a systematic way for organizing metabolic networks into such pathways. While decomposition using elementary flux modes has proven to be a powerful tool for understanding and manipulating cellular metabolism, its utility, however, is severely limited since the number of modes in a network increases exponentially with its size.


Here, we present a new method for decomposition of metabolic flux distributions into elementary flux modes. Our method can easily operate on large, genome-scale networks since it does not require all relevant modes of the metabolic network to be generated. We illustrate the utility of our method for metabolic engineering of Escherichia coli and for understanding the survival of Mycobacterium tuberculosis (MTB) during infection.


Our method can achieve computational time improvements exceeding 2000-fold and requires only several seconds to generate elementary mode decompositions on genome-scale networks. These improvements arise from not having to generate all relevant elementary modes prior to initiating the decomposition. The decompositions from our method are useful for understanding complex flux distributions and debugging genome-scale models.