A note on stochastic modeling of biological systems: automatic generation of an optimized Gillespie algorithm
Signaling pathways and gene regulatory networks (GRNs) play a central role in the signal trans-duction and regulation of biochemical processes occurring within the cellular environment. Under-standing their mechanisms and dynamics is of major interest in various areas of life sciences and biological sciences. For example controlling stem cell fate decision requires a comprehension of the dynamical behavior of the networks involved in stem cell differentiation and pluripotency mainte-nance. In addition to analytical mathematical methods which are applicable for small or medium sized systems, there are many computational approaches to model and analyze the behavior of larger systems. However, from a dynamical point of view, modeling a combination of signaling pathways and GRNs present several challenges. Indeed, in addition to being of large dimensionality, these systems have specific dynamical features. Among the most commonly encountered is that the signal transduction controlled by the signaling pathways occurs at a different time scale than the transcription and translation processes. Also, stochasticity is known to strongly impact the regulation of gene expression. In this paper, we describe a simple implementation of an optimized version of the Gille-spie algorithm for simulating relatively large biological networks which include delayed reactions. The implementation presented herein comes with a script for automatically generating the different data structures and source files of the algorithm using standardized input files.\n\nCode availabilityThe Fortran90 implementation of the code and the R script described here as well as the tutorial with practical instructions are stored on the following github repository qvhaelen/ typhon