State Function-Based Correction: A Simple and Efficient Free-Energy Correction Algorithm for Large-Scale Relative...

Area of Science:

• Computational Chemistry
• Drug Discovery
• Molecular Modeling

Background:

• Free-energy perturbation-based relative binding free-energy (FEP-RBFE) calculations are crucial in drug discovery.
• Inherent computational errors necessitate corrections for improved prediction accuracy.
• Traditional cycle-based correction methods face scalability issues with increasing network sizes.

Purpose of the Study:

• To introduce an efficient and scalable algorithm for correcting FEP-RBFE calculations.
• To overcome the computational bottlenecks of traditional methods in large-scale drug discovery screens.
• To enhance the accuracy and applicability of FEP-RBFE for high-throughput lead optimization.

Main Methods:

• Developed the State Function-based Correction (SFC) algorithm, leveraging the state function property of free energy.
• SFC avoids the need for computationally expensive cycle identification in perturbation graphs.
• The algorithm's computational cost scales linearly with the number of edges and molecules (O(P × N)).

Main Results:

• SFC demonstrates consistent computational efficiency across increasing graph sizes, unlike graph-based methods.
• The algorithm efficiently handles large perturbation networks, including those with up to 50,000 molecules.
• Incorporation of uncertainty-aware weighting further improves correction performance.

Conclusions:

• SFC offers a computationally efficient and scalable solution for correcting FEP-RBFE calculations.
• The method is well-suited for high-throughput FEP-RBFE applications in modern drug discovery.
• SFC enhances the reliability of FEP-RBFE for lead optimization processes.