Suitability of Genetic Algorithm Generated Models to Simulate Atrial Fibrillation and K⁺Channel Blockades

Channel modifications resulting from atrial fibrillation (AF) have received a great deal of attention over the last decade. Mathematical models can be used to help understand the significance of these changes. These models can be used to predict the responses of specific channel-blocker drugs on normal action potentials (NAPs) and action potentials (APs) present during chronic AF (AFAP). Unfortunately, to date, models are "average representations" of APs, but AP morphology varies significantly through the atria. To account for this natural heterogeneity, which plays a very important role in determining the nature of AF, we previously presented a genetic algorithm (GA) to automatically fit the conductance parameters of atrial model APs based upon experimentally measured APs. In this study, three automatically produced models from different canines were used to investigate the suitability of this technique in assessing the effects of AF and drug-related channel modifications.