Atrial fibrillation drug structural analysis and predictive modeling employing machine learning methods

Shumaila Noreen, Shahid Zaman, Shazia Nawaz, Neveen Ali Eshtewy,
Atrial fibrillation drug structural analysis and predictive modeling employing machine learning methods,
Next Research,
Volume 2, Issue 4,
2025,
100866,
ISSN 3050-4759,
https://doi.org/10.1016/j.nexres.2025.100866.
(https://www.sciencedirect.com/science/article/pii/S305047592500733X)
Abstract: Quantitative structure-property relationship (QSPR) modeling offers a powerful approach to predict pharmacological and physicochemical properties of drugs, yet its application in cardiovascular pharmacology remains limited. With a focus on beta-blockers, we report the first machine learning-based QSPR analysis of atrial fibrillation medications in this study. For 11 chosen medications, structural characteristics were evaluated using topological indices and molecular descriptors. Thirty percent of the dataset was used for testing and seventy percent was used for training regression models, such as Linear regression, Random Forest and Decision Tree regression. The normalized root mean square error (NRMSE), coefficient of determination (R2) and Mean Absolute error were used to assess the model’s performance. molecule weight predictions were somewhat less accurate, indicating the sensitivity of topological indices to molecule size and shape, but molar volume predictions obtained R2>0.90 with low NRMSE. These findings show that molecular topology can be efficiently encoded by QSPR models to forecast characteristics associated with drug activity. The presented work shows the possibility of integrating cheminformatics and machine learning in cardiovascular pharmacology by offering new insights into predictive modeling of medications for atrial fibrillation. In order to enhance predicted accuracy and biological relevance, our findings set the stage for future research that uses bigger datasets, three-dimensional descriptors, and experimental verification.
Keywords: QSPR analysis; Topological indices; Linear regression; Random Forest; Decision Tree Regression