Methodology article

Exhaustive identification of steady state cycles in large stoichiometric networks

Jeremiah Wright^1,2 and Andreas Wagner^1,2,3,4

¹Department of Biochemistry, University of Zurich, Zurich, Switzerland

²Swiss Institute of Bioinformatics, Lausanne, Switzerland

³Sante Fe Institute, Sante Fe, New Mexico, USA

⁴Department of Biology, The University of New Mexico, Albuquerque, New Mexico, USA

author email corresponding author email

BMC Systems Biology 2008, 2:61doi:10.1186/1752-0509-2-61

Published:	11 July 2008

Abstract

Background

Identifying cyclic pathways in chemical reaction networks is important, because such cycles may indicate in silico violation of energy conservation, or the existence of feedback in vivo. Unfortunately, our ability to identify cycles in stoichiometric networks, such as signal transduction and genome-scale metabolic networks, has been hampered by the computational complexity of the methods currently used.

Results

We describe a new algorithm for the identification of cycles in stoichiometric networks, and we compare its performance to two others by exhaustively identifying the cycles contained in the genome-scale metabolic networks of H. pylori, M. barkeri, E. coli, and S. cerevisiae. Our algorithm can substantially decrease both the execution time and maximum memory usage in comparison to the two previous algorithms.

Conclusion

The algorithm we describe improves our ability to study large, real-world, biochemical reaction networks, although additional methodological improvements are desirable.