A modified tau-leaping method for gene regulatory networks
Stochastic modeling is valuable in investigating rare phenomena, with simulation in particular serving as a computationally expensive yet accessible tool. Developments in the past several decades have accelerated simulation algorithms by reducing the cost per simulation step, in the case of exact methods, or by characterizing and addressing stiffness in the stochastic setting, in the case of approximate methods. However, existing algorithms are ill-equipped to deal with certain motifs in gene regulatory networks. Motivated by a gene regulatory model explaining breast cancer heterogeneity, this talk presents a modified tau-leaping method to accelerate models with bounded compartments.