Divide-and-Conquer Approach to the Parallel Computation of Elementary Flux Modes in Metabolic Networks

Jevremovic, Dimitrije; Boley, Daniel; Sosa, Carlos

doi:10.1109/ipdps.2011.188

Cited by 13 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The order in which reactions are processed greatly affects the performance of the Double Description method. A common strategy is to sort reaction rows according to the number of possible combination between the positive and negative elements presented within rows of the null space matrix [ 16 ],[ 29 ]. In 100 randomly ordered networks based on the (smallest) model subaa , computation time using the depth-first algorithm ranged from 310 seconds to almost 5000 seconds (Figure 4 A).…”

Section: Resultsmentioning

confidence: 99%

“…These approaches are based on the classical Nullspace approach [ 9 ],[ 14 ]. Recent elegant strategies for problem sub-division [ 15 ],[ 16 ] have lessened the physical memory load from an explosion of intermediate flux modes. Load balancing across and the coordination of computing nodes remain to be significant issues, which limit scalability by problem sub-division.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A depth-first search algorithm to compute elementary flux modes by linear programming

Quek

Nielsen

2014

BMC Syst Biol

View full text Add to dashboard Cite

BackgroundThe decomposition of complex metabolic networks into elementary flux modes (EFMs) provides a useful framework for exploring reaction interactions systematically. Generating a complete set of EFMs for large-scale models, however, is near impossible. Even for moderately-sized models (<400 reactions), existing approaches based on the Double Description method must iterate through a large number of combinatorial candidates, thus imposing an immense processor and memory demand.ResultsBased on an alternative elementarity test, we developed a depth-first search algorithm using linear programming (LP) to enumerate EFMs in an exhaustive fashion. Constraints can be introduced to directly generate a subset of EFMs satisfying the set of constraints. The depth-first search algorithm has a constant memory overhead. Using flux constraints, a large LP problem can be massively divided and parallelized into independent sub-jobs for deployment into computing clusters. Since the sub-jobs do not overlap, the approach scales to utilize all available computing nodes with minimal coordination overhead or memory limitations.ConclusionsThe speed of the algorithm was comparable to efmtool, a mainstream Double Description method, when enumerating all EFMs; the attrition power gained from performing flux feasibility tests offsets the increased computational demand of running an LP solver. Unlike the Double Description method, the algorithm enables accelerated enumeration of all EFMs satisfying a set of constraints.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A depth-first search algorithm to compute elementary flux modes by linear programming

Quek

Nielsen

2014

BMC Syst Biol

View full text Add to dashboard Cite

show abstract

“…The second ones use the stoichiometric data to do calculations during the process. Linear Programming and Null-Space Algorithm [20] are some of the mathematical strategies applied to find pathways, mainly solving the system of linear equations proposed by the stoichiometric matrix.…”

Section: Methodsmentioning

confidence: 99%

Improving the EFMs quality by augmenting their representativeness in LP methods

et al. 2018

View full text Add to dashboard Cite

BackgroundAlthough cellular metabolism has been widely studied, its fully comprehension is still a challenge. A main tool for this study is the analysis of meaningful pieces of knowledge called modes and, in particular, specially interesting classes of modes such as pathways and Elementary Flux Modes (EFMs). Its study often has to deal with issues such as the appearance of infeasibilities or the difficulty of finding representative enough sets of modes that are free of repetitions. Mode extraction methods usually incorporate strategies devoted to mitigate this phenomena but they still get a high ratio of repetitions in the set of solutions.ResultsThis paper presents a proposal to improve the representativeness of the full set of metabolic reactions in the set of computed modes by penalizing the eventual high frequency of occurrence of some reactions during the extraction. This strategy can be applied to any linear programming based extraction existent method.ConclusionsOur strategy enhances the quality of a set of extracted EFMs favouring the presence of every reaction in it and improving the efficiency by mitigating the occurrence of repeated solutions. The new proposed strategy can complement other EFMs extraction methods based on linear programming. The obtained solutions are more likely to be diverse using less computing effort and improving the efficiency of the extraction.

show abstract

“…Efficient solutions to this problem have been considered in the divide-and-conquer approach [24,25] by rearranging the constraints in an echelon form so that the constraints containing only the desired reactions appear at the bottom. To define the constraints in our case, we consider first the hypergraph

{scriptR}_{T}

that is formed only by heterologous reactions.…”

Section: Definitionsmentioning

confidence: 99%

Enumerating metabolic pathways for the production of heterologous target chemicals in chassis organisms

et al. 2012

View full text Add to dashboard Cite

BackgroundWe consider the possibility of engineering metabolic pathways in a chassis organism in order to synthesize novel target compounds that are heterologous to the chassis. For this purpose, we model metabolic networks through hypergraphs where reactions are represented by hyperarcs. Each hyperarc represents an enzyme-catalyzed reaction that transforms set of substrates compounds into product compounds. We follow a retrosynthetic approach in order to search in the metabolic space (hypergraphs) for pathways (hyperpaths) linking the target compounds to a source set of compounds.ResultsTo select the best pathways to engineer, we have developed an objective function that computes the cost of inserting a heterologous pathway in a given chassis organism. In order to find minimum-cost pathways, we propose in this paper two methods based on steady state analysis and network topology that are to the best of our knowledge, the first to enumerate all possible heterologous pathways linking a target compounds to a source set of compounds. In the context of metabolic engineering, the source set is composed of all naturally produced chassis compounds (endogenuous chassis metabolites) and the target set can be any compound of the chemical space. We also provide an algorithm for identifying precursors which can be supplied to the growth media in order to increase the number of ways to synthesize specific target compounds.ConclusionsWe find the topological approach to be faster by several orders of magnitude than the steady state approach. Yet both methods are generally scalable in time with the number of pathways in the metabolic network. Therefore this work provides a powerful tool for pathway enumeration with direct application to biosynthetic pathway design.

show abstract

Divide-and-Conquer Approach to the Parallel Computation of Elementary Flux Modes in Metabolic Networks

Cited by 13 publications

References 27 publications

A depth-first search algorithm to compute elementary flux modes by linear programming

A depth-first search algorithm to compute elementary flux modes by linear programming

Improving the EFMs quality by augmenting their representativeness in LP methods

Enumerating metabolic pathways for the production of heterologous target chemicals in chassis organisms

Contact Info

Product

Resources

About