A Review of Computational Methods in Predicting hERG Channel Blockers

Abstract

Inhibition of the human ether‐à‐go‐go‐related gene (hERG) ion channel may lead to prolonged QT interval and even fatal ventricular arrhythmia. Therefore, evaluating the hERG liability has become a mandatory requirement for drug development and drug safety process. However, standard experimental assays are complex, expensive and time‐consuming. Various in silico tools have been developed to identify potential hERG blockers and reduce in advance the risk of cardiotoxicity‐related attritions during the early drug discovery phase. In this review, we first outlined the structure and gating properties of the hERG channel briefly. Then analyzed the advantages of computational technology, and comprehensively summarized the in silico methods of hERG‐related cardiotoxicity, including ligand‐based and structure‐based methods. Additionally, the current predictive models in prediction of hERG blockage were emphatically reviewed. Finally, issues involved in the current computational methods, as well as the future direction for developing reliable models of hERG channel blockers, are discussed.

Mitigating hERG Liability of Toll‐Like Receptor 9 and 7 Antagonists through Structure‐Based Design

“…However, assessment of these compounds in hERG assay showed most of the potent TLR7/9 antagonists suffer from hERG liability, with experimental hERG IC 50 value <10 μM, rendering the molecules unsuitable for further pre‐clinical development. There have been recent reports of employing in‐silico tools including ligand‐based and structure‐based methods for identifying potential hERG blockers to reduce cost, time and resources [35] . Structure‐based methods are considered to be advantageous because of their wider applicability as compared to the ligand‐based approach, which requires a significantly large number of diverse molecules in the training set.…”