Figure 4

Genetic distance to a genotype network is negatively correlated with a phenotype's penetrance. (a) The leftmost circle represents a genotype G, where an alternative phenotype $s_{\infty }^{new}$ has a given penetrance (red sector). We quantified the penetrance of $s_{\infty }^{new}$ in G as the fraction of single-gene perturbations of the initial condition s0 that produce $s_{\infty }^{new}$. We determined genetic distance as the smallest number of mutations required to reach $s_{\infty }^{new}$'s genotype network (rightmost circle). (b) The horizontal axis shows the penetrance of a phenotype $s_{\infty }^{new}$. The vertical axis shows the distance to $s_{\infty }^{new}$'s genotype network. A circle's area is proportional to the number of data points in each penetrance and mutational distance category. The panel is based on 10⁴ circuit genotypes with N = 16 genes, c ≈ 0.35, and d = 0.25. Penetrance and genetic distance to the new genotype network are negatively associated (Spearman's ρ = -0.293; p < 2.2 × 10^-16). The inset shows that the fraction of genotypes that are neighbors of the new genotype network increases with penetrance.