Accelerating Alchemical Free Energy Prediction Using a Multistate Method: Application to Multiple Kinases

Area of Science:

• Computational chemistry
• Molecular dynamics simulations
• Drug discovery

Background:

• Alchemical free-energy methods aid in optimizing small organic molecules for improved protein binding affinity.
• Pairwise free-energy calculations become computationally prohibitive as the number of molecules to assess increases.

Purpose of the Study:

• To introduce and validate replica-exchange enveloping distribution sampling (RE-EDS) for calculating multi-ligand alchemical free-energy differences.
• To demonstrate the general applicability of RE-EDS in prospective drug design campaigns.

Main Methods:

• Developed replica-exchange enveloping distribution sampling (RE-EDS), a pathway-independent multistate method.
• Applied RE-EDS to a set of four kinases and 42 corresponding inhibitors.
• Performed single molecular dynamics (MD) simulations to calculate free-energy differences between multiple ligands simultaneously.

Main Results:

• RE-EDS successfully modeled up to 13 ligands concurrently for the studied kinase targets.
• High sampling efficiency was achieved with the RE-EDS method.
• A substantial reduction in computational cost was observed compared to traditional pairwise methods.

Conclusions:

• RE-EDS offers a computationally efficient approach for alchemical free-energy calculations involving multiple ligands.
• The method shows promise for accelerating lead optimization in drug design campaigns.
• RE-EDS provides a scalable solution to the combinatorial challenge of assessing numerous potential drug candidates.