Binding Free Energies without Alchemy.

Area of Science:

• Computational chemistry
• Molecular modeling
• Drug discovery

Background:

• Absolute Binding Free Energy (ABFE) methods offer high accuracy in predicting protein-ligand binding affinities.
• Current ABFE methods often require numerous simulations of alchemically modified intermediate states, limiting their practical application.
• There is a need for more efficient computational techniques for binding affinity prediction.

Purpose of the Study:

• To introduce Direct Binding Free Energy (DBFE), a novel end-state ABFE method.
• To evaluate the performance of DBFE compared to existing methods.
• To develop a more computationally efficient approach for predicting binding affinities suitable for virtual screening.

Main Methods:

• Developed DBFE, an end-state ABFE method utilizing implicit solvent.
• DBFE eliminates the need for alchemical intermediate states.
• Receptor and ligand simulations can be precomputed and amortized.

Main Results:

• DBFE demonstrated superior performance over OBC2 double decoupling on a host-guest benchmark.
• DBFE achieved performance comparable to OBC2 MM/GBSA on a protein-ligand benchmark.
• The method requires only one complex simulation per ligand, significantly reducing computational cost compared to double decoupling.

Conclusions:

• DBFE is an accurate and efficient computational method for predicting protein-ligand binding affinities.
• The reduced simulation requirements make DBFE a promising candidate for large-scale virtual screening workflows.
• The open-source release of the DBFE code facilitates its adoption and further development.