Computing Solvation Free Energies of Small Molecules with Experimental Accuracy

Free energies play a central role in characterizing the behavior of chemical systems and are among the most important quantities that can be calculated by molecular dynamics simulations. Solvation free energies in various organic solvents, in particular, are well-studied physicochemical properties of drug-like molecules and are commonly used to assess and optimize the accuracy of nonbonded parameters in empirical force fields and also as a fast-to-compute surrogate of performance for protein-ligand binding free energy estimation. Machine learned potentials (MLPs) show great promise as more accurate alternatives to empirical force fields but are not readily decomposed into physically motivated functional forms, which has thus far rendered them incompatible with standard alchemical free energy methods that manipulate individual pairwise interaction terms. However, since the accuracy of free energy calculations is highly sensitive to the force field, this is a key area in which MLPs have the potential to address the shortcomings of empirical force fields. In this work, we introduce an efficient alchemical free energy protocol that enables calculations of rigorous free energy differences in condensed phase systems modeled entirely by MLPs. Using a pretrained, transferable, alchemically equipped MLP model, we demonstrate subchemical accuracy for the solvation free energies of a wide range of organic molecules.