We now integrate equations (1-3) in Matlab using the following code. First camp.m:
global nu sigma k kt L q h nu = 0.1; %nu = 0.04; sigma = 1.2; k = 0.4; kt = 0.4; L = 10^6; q = 100; h = 10; [T,Y] = ode23('func_camp',[0,2500],[92.366,10,2]); figure(1); hold on; plot(T,Y(:,1),'b'); plot(T,Y(:,2),'r'); plot(T,Y(:,3),'g'); xlabel('t'); ylabel('\alpha,\beta,\gamma');Text version of this program
function dy = func_camp(t,y) global nu sigma k kt L q h a = y(1); b = y(2); g = y(3); dalpha = nu - sigma*phi(a,g); dbeta = q*sigma*phi(a,g) - kt*b; dgamma = kt*b/h - k*g; dy = [dalpha;dbeta;dgamma];Text version of this program
function r = phi(alpha,gamma) global nu sigma k kt L q h r = (alpha*(1+alpha)*(1+gamma)^2)/(L + ((1+alpha)^2)*((1+gamma)^2));Text version of this program
The output from running camp.m is shown in Figure 2 for . Here there are sustained, stable oscillations in the three variables.
If instead we take , corresponding to a lower rate of synthesis of ATP, we find that sustained oscillations are not possible. However, it is still possible to get a single spike of cAMP concentration, as shown in Figure 3. Note that this plot is for the initial conditions . Try varying these initial conditions to see how ``easy'' it is to get such a single spike of cAMP for this model.