Linear Residual Network Modeling for Anti-HIV-1 Activity Prediction and Docking-Validated Design of...

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Drug Discovery

Background:

• Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are crucial in HIV-1 treatment.
• Predicting the activity of novel NNRTIs is essential for efficient drug development.
• Biphenyl-DAPY scaffolds represent a promising class of NNRTIs.

Purpose of the Study:

• To develop and validate quantitative structure-activity relationship (QSAR) models for predicting anti-HIV-1 activity of biphenyl-DAPY-based NNRTIs.
• To identify the most predictive model for guiding the rational design of novel NNRTIs.
• To computationally assess the binding affinity, pharmacokinetic properties, and dynamic stability of designed NNRTIs.

Main Methods:

• Quantitative Structure-Activity Relationship (QSAR) analysis using heuristic method (HM) for descriptor selection.
• Development and validation of predictive models including Support Vector Regression (SVR) and Linear Residual Network (LRNet).
• Molecular docking, in silico ADMET profiling, and molecular dynamics (MD) simulations for external validation and lead optimization.

Main Results:

• The Linear Residual Network (LRNet) model demonstrated superior predictive performance with high R2 values on training and test sets (0.8838 and 0.9026, respectively).
• LRNet showed excellent generalization on an independent external dataset, confirming its robustness.
• Designed novel compounds exhibited favorable binding modes with the 1RT2 protein, good pharmacokinetic profiles, and stable dynamic interactions within the binding pocket.

Conclusions:

• The developed QSAR framework, particularly the LRNet model, provides a reliable computational tool for predicting the anti-HIV-1 activity of biphenyl-DAPY-based NNRTIs.
• This approach facilitates the rational design and optimization of novel NNRTIs with enhanced therapeutic potential.
• The study highlights the utility of integrated computational methods in accelerating anti-HIV drug discovery efforts.