Cells with the same DNA can exhibit different phenotypes, thanks in part to the nonlinear dynamics of gene regulation: feedback loops in regulatory networks give rise to multiple steady-state attractors, which correspond to alternative states of gene expression. However, noisy gene expression can enable spontaneous transitions between these states. This noise-induced switching is thought to underlie critical cellular processes, including developmental fate-decisions, phenotypic plasticity in fluctuating environments, and even carcinogenesis.

This talk will discuss computational approaches that shed light on the dynamics of spontaneous switching in multi-stable gene networks. Numerical methods will be presented that tackle two challenges: the rare-event problem (switching between gene states occurs rarely, making it difficult to achieve proper sampling), and the curse of dimensionality (switching requires coordinated changes involving many species in the network). These approaches reveal how fluctuations both in occupancies of DNA regulatory sites and protein products drive switching events in common gene network motifs.