ABSTRACT In this study, a three‐dimensional quantitative structure–activity relationship (3D‐QSAR) analysis was performed on 38 quinazoline derivatives as GLP inhibitors. The Comparative Molecular Similarity Indices Analysis (CoMSIA) model showed strong internal consistency (Q 2 = 0. 515, r 2 = 1. 000) and good external predictability (K = 0. 959, K′ = 1. 038, t/t′ = –0. 338, R 2 m = 0. 98). A virtual screening of a 51, 369‐compound quinazoline library was conducted using the CoMSIA model and molecular docking to identify potential GLP inhibitors. The newly predicted compounds exhibited enhanced activity compared to the known actives BIX, E72, and UNC0224. Their ADMET properties were also evaluated, and the top three compounds (Q1, Q2, and Q3), which demonstrated superior pharmacokinetic profiles, were selected as promising GLP inhibitor candidates. Molecular dynamics (MD) simulations were used to assess the binding stability of the three compounds in the active site of 3RJW (i. e. , 3RJWQ1, 3RJWQ2, and 3RJWQ3). The results showed good stability, with a low root‐mean‐square deviation (RMSD) of lower than 0. 7 nm. The relative binding free energy (ΔG bind) values obtained from MM‐GBSA and MM‐PBSA analyses for mentioned complexes are in consistent with the total scores obtained from docking analysis, and support the reliability of the results. Per‐residue energy decomposition analysis highlighted four key amino acids—PHE1087, PHE1158, ASP1090, and PRO1121—as critical contributors to binding affinity. Furthermore, the results of topological analysis of electron density using the Quantum Theory of Atoms in Molecules (QTAIM) align with the molecular docking results, further confirming the accuracy and reliability of the computational approach.
Baluch et al. (Fri,) studied this question.