Model Comparison. <i>χ</i>² is the the weighted sum of squared residuals, AIC is the Akaike Information Criterion score, is the minimum <i>χ</i>² value amongst the models, AIC_min is the minimum AIC value amongst the models, and <i>n</i> is the number of fitted model parameters.

The AIC is an estimate of the information lost by using the model to represent the data and accounts for the fitting benefit of having more parameters; it decreases as the fit to data improves and increases with the number of parameters in the model. Here, effectively, AIC = N ln(χ2) + 2m, where N is the number of data values and m is the number of non-zero model parameters plus the number of fitted initial conditions [60]. As a rule of thumb, a difference in AIC of more than 10 between two models is strong evidence in favor of the model with the lower AIC score [61]. By both χ2 and AIC measures, the AABSM clearly describes both sets of time-course data best.