Related Experiment Video
Updated: May 16, 2026
Identification and Classification of Position-specific GABAA Receptor Subunit Missense Variants for Their Role In Hippocampal Pyramidal Neurons
Published on: June 6, 2025
Towards in silico identification of the human ether-a-go-go-related gene channel blockers: discriminative vs.
N Kireeva1, S L Kuznetsov, A A Bykov
1Frumkin Institute of Physical Chemistry and Electrochemistry RAS, Moscow, Russia. nkireeva@gmail.com
Insights
This study developed computational models to predict compounds blocking HERG channels, which cause long QT syndrome. Generative Topographic Maps showed promise for applicability domain analysis and generating probability descriptors.
Area of Science:
- Pharmacology
- Computational Chemistry
- Cardiology
Background:
- HERG potassium channels are crucial for cardiac electrical activity.
- Blockade of HERG channels can lead to fatal long QT syndrome.
- Diverse drug classes can inhibit HERG channels.
Purpose of the Study:
- To develop and compare generative (Generative Topographic Maps) and discriminative (Support Vector Machines) models for in silico classification of HERG channel-blocking compounds.
- To evaluate the utility of Generative Topographic Maps for applicability domain analysis and probability-based descriptor generation.
- To compare the predictive performance of developed models with existing studies.
Main Methods:
- Utilized Generative Topographic Maps (GTM) and Support Vector Machines (SVM) for compound classification.
- Employed various molecular descriptors to train and test the classification models.
- Assessed model performance and compared results with previously published studies on the same dataset.
Main Results:
- Both GTM and SVM models successfully classified compounds as active or inactive HERG channel blockers.
- Generative Topographic Maps demonstrated effectiveness for applicability domain assessment.
- Novel probability-based descriptors were generated using GTM, showing potential for enhanced predictive power.
Conclusions:
- Computational models, particularly GTM, offer a valuable approach for predicting HERG channel blockers.
- GTM provides a novel method for applicability domain analysis and generating predictive descriptors in drug safety assessment.
- The study highlights the importance of in silico methods for identifying compounds that may cause long QT syndrome.
Abstract:
HERG potassium channels have a critical role in the normal electrical activity of the heart. The blockade of hERG channels in heart cells can result in a potentially fatal disorder called long QT syndrome. HERG channels can be blocked by compounds with diverse structures belonging to several drug classes. Presented herein are generative (Generative Topographic Maps) and discriminative (Support Vector Machines) classification models to categorize the compounds in silico into active and inactive classes by using different types of descriptors. The predictive performance of discriminative and generative classification models has been compared. Here, the possibility of using Generative Topographic Maps as an approach for applicability domain analysis and to generate probability-based descriptors was demonstrated to our knowledge for the first time. Comparison of obtained results with the models developed by other teams on the same data set has been performed.
