Multiscale prediction of drug-induced QT prolongation

This server predicts the changes on the Action Potential (AP) produced when a compound (eg. a drug) blocks the potassium channels hERG and KCNQ1. These changes can appear in the electrocardiogram (ECG) as an elongation of the QT segment, which has been associated to an increased risk of ventricular arrhythmia (torsade de pointes) and sudden death.

The predictions provided here go beyond the hERG binding affinity prediction and show the result of applying a multiscale protocol which first predicts the binding affinities of the compound for two receptors and then simulates the effect on the Action Potential at diverse conditions. The method used is described in further detail here.

Compounds can be submitted in SMILES format. Some example molecules in SMILES format are provided here.

Submit your molecule in SMILES format

Please enable Java and JavaScritpt in your machine

Insert SMILES code:

Insert file in SMILES:


This server is available for academic purposes only. The Computer-Assisted Drug Design Lab declines any responsibility for correctness, completeness or quality of the predictions provided.


This server has been developed by Cristian Obiol-Pardo and Manuel Pastor at the Computer-Assisted Drug Design Lab IMIM, GRIB, Universitat Pompeu Fabra (UPF) in collaboration with the “Grupo de Bioelectronica I3BH”, Universidad Politecnica de Valencia (UPV). It is based on our article “A multiscale simulation system for the prediction of drug-induced cardiotoxicity”, link here.

Project funded by the Virtual Physiological Human Network of Excellence(VPH-NoE).

JME molecular editor kindly provided by Peter Ertl, Novartis Pharma.