a Reaction network for the CGM that is obtained by adding v-data between the LSRs G and GA and retains 17 reactions (reactions in dark colour) (CGM-bi)
b Fit of the data predicted by CGM-bi to the experimental data and comparison with the detailed (base) model
c The CGM predicts four distinct LSRs on the Z-plot as well as the v-plot
Black arrow indicates [RGS] suitable for biphasic response in Z upon increase in [R_active]

a Reaction network for the CGM that is obtained by adding the constraint that the reactions A₊₅ and A₋₆ must be retained. This CGM also retains 17 reactions in total
b Fit of the data predicted by CGM-sig to the experimental data and comparison with the detailed model
c Similar to CGM-bi, this CGM predicts four distinct LSRs on the Z and v plots

a Reaction network for the detailed model of the GTPase-cycle module [23]. G: G-protein, R: agonist-bound active GPCR, A: GAP, T: GTP, D: GDP. G proteins bind T, hydrolyse it to D, and D dissociates; in each state, G proteins reversibly bind R (which accelerates the GDP/GTP exchange) or GAP (which accelerates GTP hydrolysis). G * denotes the active form of G protein. Free GTP, GDP and phosphate (P_i) are not shown for simplicity. Each reaction is labelled with its name, where Ai (i = 1, 2, 3 and so on) denotes an exchange (association or dissociation) of GAP, Ri denotes an exchange of the GPCR, Ti denotes an exchange of the GTP, Pi denotes an exchange of a phosphate (GTP hydrolysis reaction) and Di denotes an exchange of the GDP. For the reactions Ri, the forward reaction is on the right side. The forward reaction is an association reaction for the Ai, Ri and Ti reactions, whereas it is a dissociation reaction for the Di and Pi reactions
b Comparison of steady-state GTPase activities derived from simulations with those from vesicle experiments. Simulations of the best parameter set are shown by regular lines and experimental data [46] by lines with the * marker. To allow for slight differences in vesicle preparations, the concentration of the receptor used in simulations (R_sim) was allowed to vary slightly from that used in vesicle experiments (R_exp). (Panel 1) Dependence on GTP concentration in the presence of a 4 μM GAP (simulation and experiment; R_exp = 3 nM and R_sim = 2.2 nM). (Panel 2) Dependence on GTP concentration in the absence of a GAP; R_exp = 3 nM and R_sim = 3 nM. (Panel 3) Dependence on GAP concentration in the presence of a 10 μM GTP; R_exp = 3 nM and R_sim = 5.4 nM. In the experiments and simulations, [G] = 10 nM. In experiments, a 1 mM (saturating) carbamoylcholine agonist was used
c The concentrations of the active receptor (R) and the GAP determine steady-state fractional G protein activation and GTPase activity. Three-dimensional logarithmic plots show the output of simulations of Z (panel 1) and v (panel 2) at various concentrations of the R and the GAP, with a 10 nM G protein and cellular concentrations of the GTP, GDP and Pi. The labelled plateaus (G, RG, RGA and GA) show the four LSRs of the GTPase cycle and indicate the proteins contributing to the regulation of G protein activity in each limit. The four LSRs essentially correspond to the four extreme paths: (1) LSR G, G → G * T → GD → G, Z = 7.6 × 10⁻³, v = 9.9 × 10⁻⁵; (2) LSR RG, RG → RG * T → RGD → RG, Z = 0.98, v = 0.012; (3) LSR RGA, RGA → RG * AT → RGAD → RGA, Z = 0.096, v = 2.4; (4) LSR GA, GA → G * AT → GAD → GA, Z = 4.6 × 10⁻⁵, v = 1.2 × 10⁻³. Among the LSRs, a large range of intermediate signalling behaviour occurs, and to transition between LSRs, changes in R or GAP concentration of 50–1000-fold are required

a Variation for CGM 1 (CGM-bi). Dominant active G-protein complex, G * T, exhibits biphasic response for increasing [R_active]
b Variation for the detailed model

For each reaction network topology, the maximum (MAX) and minimum (MIN) value of a parameter is calculated across good parameter-value sets which fit the experimental data well [7]. The fit-error cutoff for CGM-bi and CGM-sig is 1.20 times the MIN fit error for that CGM; for the detailed model, the corresponding factor is ∼1.22. These ranges are displayed as vertical bars with the MAX and MIN marked at the ends. Only the parameters that are not fixed and are retained in at least one of the CGMs are shown. In each parameter group, the bars from left to right are for the detailed model, CGM-bi and CGM-sig, respectively. On each bar, the value of the parameter listed in Table 1 is also shown with a labelled/filled mark of the same shape. For the parameter T₊₁, the values in the CGMs closest to the range of values in the detailed model are labelled with a ‘ * ’ mark