Best practices for constructing, preparing, and evaluating protein-ligand binding affinity benchmarks

David F Hahn, Christopher I Bayly, Hannah E Bruce Macdonald, John D Chodera, Antonia SJS Mey, David L Mobley, Laura Perez Benito, Christina EM Schindler, Gary Tresadern, Gregory L Warren
Preprint ahead of publication: [arXiv] [GitHub]

This living best practices paper for the Living Journal of Computational Molecular Sciences describes the current community consensus in how to curate experimental benchmark data for assessing predictive affinity models for drug discovery, how to prepare these systems for affinity calculations, and how to assess the results to compare performance.

Towards chemical accuracy for alchemical free energy calculations with hybrid physics-based machine learning / molecular mechanics potentials

Dominic A. Rufa, Hannah E. Bruce Macdonald, Josh Fass, Marcus Wieder, Patrick B. Grinaway, Adrian E. Roitberg, Olexandr Isayev, and John D. Chodera.
Preprint ahead of submission.
[bioRxiv] [GitHub]

In this first use of hybrid machine learning / molecular mechanics (ML/MM) potentials for alchemical free energy calculations, we demonstrate how the improved modeling of intramolecular ligand energetics offered by the quantum machine learning potential ANI-2x can significantly improve the accuracy in predicting kinase inhibitor binding free energy by reducing the error from 0.97~kcal/mol to 0.47~kcal/mol, which could drastically reduce the number of compounds that must be synthesized in lead optimization campaigns for minimal additional computational cost.