Stochastic Multistability from Simple Network Motifs

Abstract

Studying stochasticity in gene regulatory networks is essential for understanding important cellular processes, such as those involved in determining cellular fate. However, studying stochasticity is a challenging task. For example, stochastic simulation algorithm is inefficient in examining rare events. Direct solution of the discrete Chemical Master Equation requires enumeration and truncation of usually enormous state space. We developed the Multi-Finite Buffer method for direct solution of discrete Chemical Master Equation (mb-dCME method), which allows the probability landscape of a large class of stochastic networks to be computed exactly. Here we use the mb-dCME method to study the probability landscape of simple network motifs. Our results show that complex behavior such as multistability can arise in very simple motifs without feedback loops and cooperativity.