Computer-aided estimation of the hERG-mediated cardiotoxicity risk of potential drug components

Abstract

The hERG potassium channel is one of the most important anti-targets determining cardiotoxicity of potential drugs. Using fragmental descriptors and artificial neural networks, the predictive models of the relationship between the structure of organic compounds and their activity with respect to hERG were built, and the structural factors affecting it were analyzed. By their predictive ability and applicability domain, these models (N = 1000, Q² = 0.77, RMSE_cv = 0.45 for affinity and N = 2886, Q² = 0.60, RMSE_cv = 0.55 for channel inhibition) are superior to the previously published models and can be used to minimize the risk of cardiotoxicity during drug development.