Fig. 2

Network diagram and simulations of the initial model of HGF-activated RhoA-NO network, including known biochemical effects of mutual antagonism. A Two-node model to demonstrate that mutual antagonism is not sufficient to induce bistability. A and B are two species with zeroth order production and first order degradation rates, ksynthA, kdegA, ksynthB, kdegB for A and B respectively. A and B antagonize each other with rate constants kiA and kiB, according to the mass action equations in Additional file 1 Text S7. B The red arrows represent the inhibitory reactions. Complete specifications of the reactions appear in Additional file 1 Tables S1 and S2. C Model simulation with 100 random initial concentrations. The model was simulated, starting with random levels of all species. The resulting trajectories of the RhoA concentration over time (plotted with a random color for each trajectory) shows convergence to two different steady states. D and E Bifurcation analysis of the initial model. The steady state NO concentration is plotted against variation in the parameter D ki1 (rate constant for ROCK inhibition of Akt phosphorylation) and E ki2 (rate constant for RhoA inhibition of eNOS mRNA). F Robustness analysis of the bistability in the initial model. Single-variable robustness analysis was performed (see Procedure 4 in Additional file 1 Methods). Each of the parameters, from k1 to k15 was varied by the specified percentage, and the system was checked for bistability. White boxes denote the presence of bistability while black boxes denote the absence of bistability. The red line highlights the unperturbed parameter set. Note that bistability was retained in some columns, particularly for parameters involved in the degradation or inactivation of NO, eNOS, and cGMP