Rational Design and Multi‐tiered Computational Evaluation of Novel Darunavir Derivatives as HIV‐1 Protease Inhibitors: an Integrated Approach Using QSAR Model, Molecular Docking, MD Simulations, DFT Analysis, and ADME Profiling

DRV is currently the most effective approved HIV-1 protease inhibitor, but after a long period of clinical treatment, multidrug-resistant HIV-1 mutations have emerged. In this paper, we initially constructed QSAR models out of 45 existing DRV analogs using 2D-QSAR, CoMFA, and CoMSIA technology. The statistical findings show that the 2D-QSAR model (q² = 0.7235, r² = 0.7910, F = 41.46, p < 0.0001), CoMFA model (q² = 0.681, SEE = 0.056, r² = 0.998, F = 1434.30), and CoMSIA/SHE models (q² = 0.839, SEE = 0.087, r² = 0.996, F = 536.48) have good predictive power. Then, four compounds with potential HIV-1 protease inhibition were obtained and combined with the validation of the above models. Moreover, ADMET was used to verify its pharmacokinetic characteristics, molecular docking was used to verify its binding ability, and kinetic simulation and binding free energy were used to verify the accuracy of the docking results. The comprehensive computational investigation results show that these newly designed molecules have excellent properties and provide potential molecular structure information for the development of novel, highly efficient, and low-toxicity HIV-1 protease-targeting inhibitors.