Transition paths across the epithelial-mesenchymal transition landscape are dictated by network logic

During development and oncogenesis, cells transition reversibly from epithelial to mesenchymal states (epithelial-mesenchymal transition; EMT). Tristable EMT can be described by a three-node gene regulatory network in which multiple inputs regulate the transcription factor ZEB. In this network, and more generally, it is not known how to choose combinatorial network logic. Here, we discover that the choice of multiplicative (AND) versus additive (OR) network logic strongly affects EMT phenotypes and leads to opposing predictions regarding factors that control EMT transition paths. We show that strong inhibition of miR-200 destabilizes the epithelial state and initiates EMT for AND logic, in agreement with experimental data. Using single-cell data, stochastic simulations and perturbation analysis, we show how these results can be used to design experiments to infer EMT network logic in live cells. We go on to analyze networks controlling cell fate decisions during embryogenesis and show that, here too, logic changes cell fate landscapes upon perturbation in important ways. Our results stress the importance of considering logic in the construction of models of regulatory networks that govern cell fate decisions.