Complex versus simple models: ion-channel cardiac toxicity prediction

Abstract

There is growing interest in applying detailed mathematical models of the heart for ion-channel related cardiac toxicity prediction. However, a debate as to whether such complex models are required exists. Here an assessment in the predictive performance between two established cardiac models, gold-standard and cardiac safety simulator, and a simple linear model Bnet was conducted. Three ion-channel data-sets were extracted from literature. Each compound was designated a cardiac risk category based on information within CredibleMeds. The predictive performance of each model within each data-set was assessed via a leave-one-out cross validation. In two of the data-sets Bnet performed equally as well as the leading cardiac model, cardiac safety simulator, both of these outperformed the gold-standard model. In the 3rd data-set, which contained the most detailed ion-channel pharmacology, Bnet outperformed both cardiac models. These results highlight the importance of benchmarking models but also encourage the development of simple models.