ecmtool: fast and memory-efficient enumeration of elementary conversion modes
Bianca Buchner, Tom J. Clement, Daan H. de Groot, and Jürgen Zanghellini
Published in Bioinformatics. Volume 39, Issue 3, March 2023, btad095.
doi: 10.1093/bioinformatics/btad095
Motivation
Characterizing all steady-state flux distributions in metabolic models remains limited to small models due to the explosion of possibilities. Often it is sufficient to look only at all possible overall conversions a cell can catalyze ignoring the details of intracellular metabolism.
Such a characterization is achieved by elementary conversion modes (ECMs), which can be conveniently computed with ecmtool. However, currently, ecmtool is memory intensive, and it cannot be aided appreciably by parallelization.
Results
We integrate mplrs - a scalable parallel vertex enumeration method - into ecmtool. This speeds up computation, drastically reduces memory requirements and enables ecmtool’s use in standard and high-performance computing environments.
We show the new capabilities by enumerating all feasible ECMs of the near-complete metabolic model of the minimal cell JCVI-syn3.0. Despite the cell’s minimal character, the model gives rise to ECMs and still contains several redundant sub-networks.
emctool: (a) Run time and (b) memory consumption of ecmtool using mplrs (squares) or polco (circles) as a function of the number of excess nutrients in the minimal medium of JCVI-syn3A.