SPONGE-FEP: An Automated Relative Binding Free Energy Calculation Accelerated by Selective Integrated Tempering...

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular modeling

Background:

• Computer-aided drug discovery (CADD) accelerates the identification and optimization of drug candidates.
• Free energy perturbation (FEP) and thermodynamic integration (TI) are crucial for predicting protein-ligand binding affinities.
• Efficient sampling is essential for accurate free energy calculations in drug design.

Purpose of the Study:

• To introduce SPONGE-FEP, an automated workflow for relative binding free energy (RBFE) calculations.
• To enhance sampling efficiency using selective integrated tempering sampling (SITS).
• To demonstrate the accuracy and performance of SPONGE-FEP in predicting binding affinities.

Main Methods:

• Implementation of SPONGE-FEP with an automated workflow.
• Utilizing selective integrated tempering sampling (SITS) for enhanced sampling.
• Performing alchemical free energy calculations, perturbation map generation, and cycle closure analysis.

Main Results:

• SPONGE-FEP significantly improves sampling efficiency for rare events.
• The method enhances the performance of RBFE predictions.
• Accuracy of SPONGE-FEP is comparable to FEP+, with rapid computation times (4 h/pair on A100 GPU).

Conclusions:

• SPONGE-FEP offers an efficient and accurate approach for RBFE calculations.
• The refined SITS method in SPONGE-FEP advances CADD capabilities.
• This tool accelerates the drug discovery pipeline by improving computational predictions.