Systems biology in drug safety assessment: Use of a recalibrated hund-rudy model to predict the effect of novel drug compounds on action potential duration

Emerging practice in drug safety assessment delivers relatively early data on the potency of compounds at a range of cardiac ion channels, but there is little consensus on how to use this data to understand human risk, particularly when a compound modifies a range of mutually compensatory electrophysiological processes. In-silico physiological modelling of the processes involved offers a method to achieve this understanding but the models in the literature are not yet widely validated in the presence of a range of pharmacological profiles. We have used a model of canine ventricular myocytes and retrained it to data from a series of action potential (AP) recordings on isolated canine myocytes. By representing inter-animal variability by using an ensemble of parameters sets, we are able to evaluate the likely impact of novel compounds on a series of virtual dogs.