The influence of assortativity on the robustness and evolvability of gene regulatory networks upon gene birth

Abstract

Gene regulatory networks (GRNs) represent the interactions between genes and gene products, which drive the gene expression patterns that produce cellular phenotypes. GRNs display a number of characteristics that are beneficial for the development and evolution of organisms. For example, they are often robust to genetic perturbation, such as mutations in regulatory regions or loss of gene function. Simultaneously, GRNs are often evolvable as these genetic perturbations are occasionally exploited to innovate novel regulatory programs. Several topological properties, such as degree distribution, are known to influence the robustness and evolvability of GRNs. Assortativity, which measures the propensity of nodes of similar connectivity to connect to one another, is a separate topological property that has recently been shown to influence the robustness of GRNs to point mutations in cis-regulatory regions. However, it remains to be seen how assortativity may influence the robustness and evolvability of GRNs to other forms of genetic perturbation, such as gene birth via duplication or de novo origination. This abstract outlines a recent publication, in which we employed a computational model of genetic regulation to investigate whether the assortativity of a GRN influences its robustness and evolvability upon gene birth. We considered GRNs to be robust if they conserved all their phenotypes (attractors) following the introduction of a new gene, and evolvable if they were able to innovate at least one novel phenotype. We found that the robustness of a GRN generally increases with increasing assortativity, while its evolvability generally decreases (Figure 1; above), and this results in an increased proportion of assortative GRNs that are simultaneously robust and evolvable (Figure 1; below). This is due to: (1) Assortative GRNs have shorter attractors, which are more likely to be conserved (Figure 2), and (2) assortative GRNs have smaller out-components, resulting in a reduced chance of innovation (Figure 3). This work extends our understanding of how the assortativity of a GRN influences its robustness and evolvability to genetic perturbation.