Efficient Formulations for Exact Stochastic Simulation of Chemical Systems

One can generate trajectories to simulate a system of chemical reactions using either Gillespie's direct method or Gibson and Bruck's next reaction method. Because one usually needs many trajectories to understand the dynamics of a system, performance is important. In this paper, we presen...

Full description

Saved in:
Bibliographic Details
Published in:IEEE/ACM transactions on computational biology and bioinformatics 2011-01, Vol.8 (1), p.27-35
Main Authors: Mauch, Sean, Stalzer, Mark
Format: Article
Language:English
Subjects:
Algorithm design and analysis
Algorithms
Biological system modeling
Biological systems
Biology and genetics
Chemical analysis
Chemical Phenomena
Computational Biology - methods
Computational complexity
Computational modeling
Computer Simulation
Fires
Genes
Kinetics
Operations research
Random variables
RNA
Simulation
Stochastic Processes
Stochastic systems
Studies
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:One can generate trajectories to simulate a system of chemical reactions using either Gillespie's direct method or Gibson and Bruck's next reaction method. Because one usually needs many trajectories to understand the dynamics of a system, performance is important. In this paper, we present new formulations of these methods that improve the computational complexity of the algorithms. We present optimized implementations, available from http://cain.sourceforge.net/>, that offer better performance than previous work. There is no single method that is best for all problems. Simple formulations often work best for systems with a small number of reactions, while some sophisticated methods offer the best performance for large problems and scale well asymptotically. We investigate the performance of each formulation on simple biological systems using a wide range of problem sizes. We also consider the numerical accuracy of the direct and the next reaction method. We have found that special precautions must be taken in order to ensure that randomness is not discarded during the course of a simulation.
ISSN:1545-5963
1557-9964
DOI:10.1109/TCBB.2009.47