Machine Learning of Free Energies in Chemical Compound Space Using Ensemble Representations: Reaching Experimental Uncertainty for Solvation

TL;DR

This paper introduces a machine learning model for predicting solvation free energies across chemical space, achieving experimental accuracy with minimal computational effort by using ensemble representations and molecular dynamics sampling.

Contribution

The authors develop a novel Free energy Machine Learning (FML) model that employs Boltzmann-averaged ensemble representations and short MD simulations, reaching experimental uncertainty levels in solvation free energy predictions.

Findings

FML prediction errors decrease with training set size, reaching 0.6 kcal/mol after 490 molecules.

FML's accuracy is comparable to state-of-the-art physics-based methods.

The model effectively analyzes solvation across 116k molecules, identifying key structural features.

Abstract

Free energies govern the behavior of soft and liquid matter, and improving their predictions could have a large impact on the development of drugs, electrolytes or homogeneous catalysts. Unfortunately, it is challenging to devise an accurate description of effects governing solvation such as hydrogen-bonding, van der Waals interactions, or conformational sampling. We present a Free energy Machine Learning (FML) model applicable throughout chemical compound space and based on a representation that employs Boltzmann averages to account for an approximated sampling of configurational space. Using the FreeSolv database, FML's out-of-sample prediction errors of experimental hydration free energies decay systematically with training set size, and experimental uncertainty (0.6 kcal/mol) is reached after training on 490 molecules (80\% of FreeSolv). Corresponding FML model errors are also on par with state-of-the art physics based approaches. To generate the input representation for a new query compound, FML requires approximate and short molecular dynamics runs. We showcase its usefulness through analysis of FML solvation free energies for 116k organic molecules (all force-field compatible molecules in QM9 database) identifying the most and least solvated systems, and rediscovering quasi-linear structure property relationships in terms of simple descriptors such as hydrogen-bond donors, number of NH or OH groups, number of oxygen atoms in hydrocarbons, and number of heavy atoms. FML's accuracy is maximal when the temperature used for the molecular dynamics simulation to generate averaged input representation samples in training is the same as for the query compounds. The sampling time for the representation converges rapidly with respect to the prediction error.