Figure 2

Validation of the model characteristics. (A) Normalized action potential (V_m), [Ca²⁺]_i transient and force amplitude at 1 Hz pacing. (B) Diastolic (closed symbols) and systolic (open symbols) [Ca²⁺]_i values as a function of pacing frequency. Simulated values (black squares) are compared with the experimental results of Ito et al. [21] (red circles) and Antoons et al. [57](green triangles). The model was able to successfully reproduce the experimentally observed minimum in the systolic [Ca²⁺]_i around 1-2 Hz [21]. (C) Simulated (black) action potential (AP) is compared to measurements by Guo et al. [58] (red), Brunet et al. [59] (green) and Petrashevskaya et al. [60] (cyan). The comparison of recorded and simulated APs (note that the APs have been shifted in time for comparison) demonstrates a good qualitative agreement with the experiments. (D) [Ca²⁺]_i transient at 1 Hz pacing. Comparison of simulated results (black) to the experimental data of Williams et al. [61] (red), Maeir et al. [20] (green) and Huser et al. [32] (cyan) demonstrates a good qualitative agreement of in silico and largely variated in vivo data. (E) Caffeine-induced [Ca²⁺]_i transients. The amplitude and decay time constants of simulated data (black) are of the same magnitude with the in vivo results of Maeir et al. [20] (red) and Li et al. [19] (green).