Key result
Neural network ensemble-based QSAR models combined with feature selection produced predictive models with very good internal cross-validation and external predictivity for hERG channel inhibition.
Population
Large and diverse set of compounds measured in a single, consistent hERG channel inhibition assay
Design
Preclinical
Authors
Loading...
May aid early cardiotoxicity screening in discovery; leaves open clinical translation pending validation.
Neural network ensemble-based QSAR models can accurately predict hERG channel inhibition, aiding in early ADME/Tox profiling in drug discovery.
Seierstad et al. (2006) studied hERG channel inhibition. QSAR models based on neural network ensembles was evaluated on Model predictivity and internal cross-validation statistics. Neural network ensemble-based QSAR models combined with feature selection produced predictive models with very good internal cross-validation and external predictivity for hERG channel inhibition.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: