Abstract
<title>Abstract</title> <p>Beta-site amyloid precursor protein cleaving enzyme 1 (BACE-1) plays a central role in amyloid-β production and remains a central therapeutic target in Alzheimer’s disease. In this study, an integrated quantitative structure activity relationship and molecular dynamics (QSAR–MD) framework was developed to identify novel and structurally stable BACE-1 inhibitors. A curated dataset of 1,532 compounds was used to construct QSAR models based on twelve molecular fingerprint classes. CDK and CDK Extended descriptors exhibited the highest predictive performance (Q²_CV = 0.78 ± 0.03; Q²_Ext = 0.79 ± 0.01), while PubChem, MACCS, Klekota_Roth, and Substructure Count also showed strong cross-validated and external predictive ability (Q²_Ext ≈ 0.76–0.78). Narrow R²–Q² gap values across all models (0.10–0.16), indicated robust generalizability. These high-confidence models were then applied to virtual screening of ZINC and NPASS libraries, identifying five top-ranked candidates. Docking and interaction profiling showed that the selected ligands formed strong hydrogen-bonding and hydrophobic interactions with key catalytic residues of BACE-1 including Asp32, Asp228, Thr72, and Thr231. Gaussian accelerated molecular dynamics (GaMD) simulations over 200 ns demonstrated that most complexes remained structurally stable, with low RMSD fluctuations, compact radius of gyration, and restricted conformational sampling. RMSD distribution analysis further supported the stable binding behavior of the top candidates. MM/GBSA calculations provided additional energetic validation, identifying NPC475168 as the most favorable binder (ΔG_total = –64.84 kcal/mol), followed by NPC313966, NPC252056, and NPC471858, all of which outperformed or matched the reference inhibitor CNP520. Overall, this study identified multiple computationally promising BACE-1 inhibitor scaffolds, with NPC475168 emerging as the promising computational lead. These findings provide a computational foundation for structural guided optimization and experimental validation of next-generation BACE-1 inhibitors in Alzheimer’s disease.</p>