The effects of a simplified model of chromatin dynamics on attractors robustness in random boolean networks with self-loops: An experimental study

Abstract

Boolean networks are currently acknowledged as a powerful model for cell dynamics phenomena. Recently, the possibility of modelling methylation mechanisms—involved in cell differentiation—in Random Boolean Networks have been discussed: methylated genes are represented in the network as nodes locked to value 0 (frozen nodes). Preliminary results show that this mechanism can reproduce dynamics with characteristics in agreement with those of cell undergoing differentiation. In a second, parallel work, the effect of nodes with self-loops in Random Boolean Networks has been studied, showing that the average number of attractors may increase with the number of self-loops, whilst the average attractor robustness tends to decrease. As these two studies are aimed at extending the applicability of Random Boolean Networks to model cell differentiation phenomena, in this work we study the combined effect of the previous two approaches. Results in simulation show that frozen nodes tend to partially dampen the effects of self-loops on attractor number and robustness. This outcome suggests that both the variants can indeed be effectively combined in Boolean models for cell differentiation.