<![CDATA[B-cell lymphoma 2 (BCL-2) is an anti-apoptotic protein implicated in the progression and chemoresistance of multiple cancers. This study integrates QSAR-based machine learning (ML) classification and molecular docking to identify potential BCL-2 inhibitors. The Gradient Boosting classifier trained on PubChem fingerprints (881 bits) achieved the best predictive performance (accuracy: 83.52%, ROC-AUC: 0.8829). External virtual screening further identified high-probability active compounds, including Beclomethasone Dipropionate (0.9880) and Ulipristal Acetate (0.9866), highlighting their potential for drug repurposing. Docking analysis showed that Ulipristal Acetate exhibited the strongest binding affinity (–8.187 kcal/mol), forming a hydrogen bond with GLY A:145 and engaging key residues within the BH3-binding pocket of BCL-2. These findings demonstrate the effectiveness of QSAR-ML–assisted virtual screening in prioritizing repurposable candidates for BCL-2 inhibition.]]>
Copyrights © 2025
