herg

This workflow includes Quantitative Structure-Activity Relationship (QSAR) models to predict the hERG-related cardiotoxicity of chemicals.

Collection of experimental protocols in common patch clamp formats for rapid characterisation of ion channels

Tool for building and manipulating Markov models of ion-channel kinetics

Code and figures for "Neural network differential equations for ion channel modelling".

Code to add a mathematical model of voltage-clamp artefacts to your ion channel simulations

Rapid characterisation of hERG channel kinetics

Code for training and validation of ion channel models fitted to sine-wave based voltage clamp data

Source code and data for reproducing "The impact of uncertainty in hERG binding mechanism on in silico predictions of drug-induced proarrhythmic risk".

Curated data deposit for the CardioSafe cardiac ion channel benchmark: labels, Tanimoto-controlled splits (tan70 / tan60), and supplementary artifacts for hERG, Nav1.5, Cav1.2, and IKs.

Master Thesis in AI focusing on TabNet in Drug Discovery

hERG cardiotoxicity prediction from ChEMBL: 11,059 compounds, scaffold-split AUROC 0.801 — and the random split that flatters it to 0.860.