Predicting Binding Free Energies in a Large Combinatorial Chemical Space Using Multisite λ Dynamics.

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular modeling

Background:

• HIV Reverse Transcriptase (HIV-RT) is a key target for antiviral therapies.
• Accurate prediction of inhibitor binding affinities is crucial for drug development.
• Conventional free energy methods face scalability challenges in exploring large chemical spaces.

Purpose of the Study:

• To demonstrate the scalability of the biasing potential replica exchange multisite λ dynamics (BP-REX MSλD) free energy method.
• To calculate binding affinities for a large set of inhibitors against HIV-RT.
• To identify novel inhibitor designs for HIV-RT with high predicted potency.

Main Methods:

• Application of the BP-REX MSλD free energy calculation method.
• Screening of 512 potential inhibitors against HIV-RT.
• Comparison of computational resources required by BP-REX MSλD versus conventional methods.

Main Results:

• The study represents the largest chemical space exploration to date using free energy methods.
• Identified 55 new potential HIV-RT inhibitor designs with predicted potency comparable to a reference compound (56 nM).
• BP-REX MSλD demonstrated an order of magnitude reduction in computational resources compared to conventional methods while maintaining precision.

Conclusions:

• BP-REX MSλD offers a highly scalable and computationally efficient approach for free energy calculations.
• This method overcomes the limitations of conventional techniques in exploring large chemical spaces.
• BP-REX MSλD significantly advances in silico drug discovery for targets like HIV-RT.