The Metropolis algorithm and the principal parameters characterizing a SA run. (A) A random modification of one of the parameters is made and the resulting fitting error is compared with the error obtained with the original parameter set. If the new error is smaller than the previous one, the new parameter value is accepted. In contrast, if the new error is larger, it is not necessarily rejected. Instead, a temperature-dependent transition probability is calculated, and if its value is greater than a random number between 0 and 1, the change is accepted. Then, the loop starts again and a new parameter to be modified is randomly chosen. (B) Dynamic adjustment of the maximal amplitude of random changes applied to the kinetic parameters of the model. The amplitude decreases when the system is cooled down to keep an acceptance ratio of ∼50%. (C) Evolution of the fitting error throughout a simulation.