Optimizing experimental designs for model selection of ion channel drug-binding mechanisms

The rapid delayed rectifier current carried by the human Ether-à-go-go-Related Gene (hERG) channel is susceptible to drug-induced reduction, which can lead to an increased risk of cardiac arrhythmia. Establishing the mechanism by which a specific drug compound binds to hERG can help reduce uncertainty when quantifying pro-arrhythmic risk. In this study, we introduce a methodology for optimizing experimental voltage protocols to produce data that enable different proposed models for the drug-binding mechanism to be distinguished. We demonstrate the performance of this methodology via a synthetic data study. If the underlying model of hERG current is known exactly, then the optimized protocols generated show noticeable improvements in our ability to select the true model when compared with a simple protocol used in previous studies. However, if the model is not known exactly, and we assume a discrepancy between the data-generating hERG model and the hERG model used in fitting the models, then the optimized protocols become less effective in determining the 'true' binding dynamics. While the introduced methodology shows promise, we must be careful to ensure that, if applied to a real data study, we have a well-calibrated model of hERG current gating.This article is part of the theme issue 'Uncertainty quantification for healthcare and biological systems (Part 1)'.