hERG-Pred allows simultaneously predicting pIC50, pKi values and classes of blockers and non-blockers based on exact and inexact values of IC50, Ki and Inhibition %.

Such multi-aspect prediction provides information for precise analysis of ligand possibility to block hERG channels.

Any potential drug can be recommended to be investigated using the created web application to evaluate the interaction with the hERG channels.

It should be mention that the quality and applicability domain of SAR models is higher than QSAR models therefore during the interpretation on the prediction results user should take it into account. If at least one SAR model predicts that a substance can be a blocker of hERG channels there is a high risk of interaction with hERG channels, therefore this substance should be tested in experimental studies to check its hERG blocking.

The risk of the hERG channels blocking increases with an increase of the number of "active" predictions by the SAR models and prediction by the QSAR models of pIC50 or pKi values higher 6.