We are currently continuing the exploration of this type B mechanism for our future work, as its novel way to differentiate has shown to provide high robustness to noise near tipping points under the right circumstances. For anyone interested in discussing, please feel free to reach out.

These different types align with the idea of Waddington's landscape as a decrease in potency, but differ in their properties, such as variance between cells, gene activity, attractor types, robustness to perturbations, and Lyapunov exponent

Type A uses chaos to grow small perturbations into distinct orbits. Type B sits on fixed points and differentiates through noise amplification, pushing a cell to both sides of a tipping point. Type C quenches and quickly falls into its final state through saddle nodes or SNIC bifurcations

In this work, we evolved GRNs with a slower timescale feedback mechanism and noise in the gene expression with the goal to differentiate to multiple cell types. In doing so, we discovered 3 distinct dynamics that can lead to cell differentiation from a dynamical systems perspective.