Topological conditions of gene regulatory networks enabling robust binary cell-fate decision-making

Embryonic development entails multi-layer branching of cell-fate decision processes; often, each branch corresponding to a binary cell-fate decision. The two phenotypes at such branching points usually have mutually exclusive gene expression patterns, driven by two master regulators that mutually inhibit each other and can activate themselves, forming a toggle switch. The dynamics of a toggle switch have been extensively studied, however, the regulatory networks underlying cell-fate decision often involve many transcription factors. We recently showed that these networks can be organized as two mutually inhibiting “teams of nodes”, such that the nodes of the same team predominantly activate themselves, while those of the opposite teams primarily inhibit each other. While the ideal cases of fully connected and consistent networks of two “teams of nodes” have been well-understood, a detailed understanding of the effect of varying levels of connectivity and inconsistency in the edges with respect to the configuration of teams is lacking. Here, we simulate networks of two-teams (TTN) under varying degrees of edge density, edge impurity and external influence to identify the conditions under which these networks can give rise to a robust binary cell-fate decision system. We find that the emergent cell-fate decision landscape is much more sensitive to inconsistent edges than to low connectivity, and that TTN generally lead to a more robust landscape than a toggle switch in the context of buffering external perturbations. Our results offer insights into the design principles of gene regulatory networks engaged in binary cell-fate decisions and can be extended to evaluate strategies for cellular reprogramming.