Related Experiment Video
Updated: Aug 14, 2026
In Silico Clinical Trials for Cardiovascular Disease
Published on: May 27, 2022
Cardiosim-Tox: an interpretable multitask deep learning QSAR platform with multimodal feature fusion for predicting
Fauzan Syarif Nursyafi1,2, Byunggyu Kang1, Junhyeok Eom1
1Department of Biomedical Engineering, Kumoh National Institute of Technology, Gumi, 39177, Republic of Korea.
Insights
Cardiosim-Tox, a deep learning platform, accurately predicts drug-induced cardiotoxicity by assessing multiple cardiac ion channels. This tool enhances early drug safety screening, reducing development attrition.
Area of Science:
- Computational chemistry and toxicology
- Pharmacology and drug safety
- Artificial intelligence in drug discovery
Background:
- Drug-induced cardiotoxicity is a major cause of drug failure, often due to cardiac ion channel blockade.
- Current computational models struggle to assess multi-channel effects, necessitating improved screening tools.
Purpose of the Study:
- To develop Cardiosim-Tox, a deep learning platform for simultaneous prediction of cardiac ion channel blockade risk and potency.
- To integrate diverse molecular features for comprehensive cardiotoxicity assessment.
Main Methods:
- Developed a modular, multi-modal deep learning platform (Cardiosim-Tox) using topological fingerprints, Mordred descriptors, and graph encodings.
- Employed both single-task (STL) and multi-task learning (MTL) configurations on a dataset of 31,816 compounds.
- Validated model performance on external datasets with domain-shift and applicability domain analyses.
Main Results:
- Multi-task learning (MTL) models consistently outperformed single-task learning (STL).
- The full multimodal configuration achieved high AUCs (0.79-0.98) for predicting blockade across hERG, Cav1.2, and Nav1.5 channels.
- MTL regression models showed excellent performance (R²=0.94), outperforming baseline models and previous cardiotoxicity predictors.
Conclusions:
- Cardiosim-Tox offers a robust and interpretable solution for integrated multi-channel cardiac safety assessment.
- The platform supports early-stage cardiotoxicity screening, aligning with the CiPA paradigm.
- The tool enhances drug development by providing reliable predictions and mechanistic insights.
Abstract:
Drug-induced cardiotoxicity, mainly driven by cardiac ion-channel blockade, remains a leading cause of drug attrition and post-market withdrawal, highlighting the need for reliable early-stage screening tools. Existing computational methods, including QSAR models, largely focus on single ion channels, limiting their ability to assess multi-channel safety profiles. To address this gap, we developed Cardiosim-Tox, a modular multi-modal deep learning platform to simultaneously predicts blockade risk (binary classification) and potency (pIC50) for hERG, Cav1.2, and Nav1.5. The framework integrates topological fingerprints, Mordred 2D/3D descriptors, and molecular graph encodings under both single-task (STL) and multi-task learning (MTL) configurations. Developed on 31,816 curated unique compounds, model performance was evaluated on three Tanimoto similarity-stratified external validation sets, with domain-shift metrics and applicability domain analyses supporting robustness. MTL models consistently outperformed STL, while the full modality combination (FP + MD + Graph) achieved AUCs of 0.93-0.98 for Cav1.2, 0.88-0.96 for Nav1.5, and 0.79-0.95 for hERG across validation sets. Regression tasks similarly favored MTL (R2 = 0.94), with multimodal configurations outperforming single-modality baselines. Further comparisons with classical machine learning models and previously developed cardiotoxicity models demonstrated the superiority of Cardiosim-Tox across all endpoints. SHAP analysis confirmed that feature-activity relationships, both at global and mechanistic levels, align with known ion-channel pharmacophores, enhancing interpretability beyond predictive accuracy. Overall, Cardiosim-Tox provides a reliable and interpretable platform for integrated multi-channel cardiac safety assessment, supporting early-stage cardiotoxicity screening in drug development in line with the CiPA paradigm. The Cardiosim-Tox web server is accessible at https://metaheart.kr/login .
