A new approach to simulating stochastic delayed systems

Farzad Fatehi; Yuliya N Kyrychko; Konstantin B Blyuss

doi:10.1016/j.mbs.2020.108327

A new approach to simulating stochastic delayed systems

Math Biosci. 2020 Apr:322:108327. doi: 10.1016/j.mbs.2020.108327. Epub 2020 Feb 28.

Authors

Farzad Fatehi¹, Yuliya N Kyrychko², Konstantin B Blyuss³

Affiliations

¹ Department of Mathematics, University of York, York YO10 5DD, UK. Electronic address: farzad.fatehichenar@york.ac.uk.
² Department of Mathematics, University of Sussex, Falmer, Brighton BN1 9QH, UK. Electronic address: Y.Kyrychko@sussex.ac.uk.
³ Department of Mathematics, University of Sussex, Falmer, Brighton BN1 9QH, UK. Electronic address: K.Blyuss@sussex.ac.uk.

PMID: 32113917
DOI: 10.1016/j.mbs.2020.108327

Abstract

In this paper we present a new method for deriving Itô stochastic delay differential equations (SDDEs) from delayed chemical master equations (DCMEs). Considering alternative formulations of SDDEs that can be derived from the same DCME, we prove that they are equivalent both in distribution, and in sample paths they produce. This allows us to formulate an algorithmic approach to deriving equivalent Itô SDDEs with a smaller number of noise variables, which increases the computational speed of simulating stochastic delayed systems. The new method is illustrated on a simple model of two interacting species and a model with bistability, and in both cases it shows excellent agreement with the results of direct stochastic simulations, while also demonstrating a much superior speed of performance.

Keywords: Delayed Fokker-Planck equation; Stochastic delay differential equation; Stochasticity; Time delay.

MeSH terms

Algorithms*
Models, Biological*
Models, Chemical*
Stochastic Processes
Time Factors