A smile is all you need: Predicting limiting activity coefficients from SMILES with natural language processing

Abstract

Knowledge of mixtures' phase equilibria is crucial in nature and technical chemistry. Phase equilibria calculations of mixtures require activity coefficients. However, experimental data on activity coefficients is often limited due to high cost of experiments. For an accurate and efficient prediction of activity coefficients, machine learning approaches have been recently developed. However, current machine learning approaches still extrapolate poorly for activity coefficients of unknown molecules. In this work, we introduce the SMILES-to-Properties-Transformer (SPT), a natural language processing network to predict binary limiting activity coefficients from SMILES codes. To overcome the limitations of available experimental data, we initially train our network on a large dataset of synthetic data sampled from COSMO-RS (10 Million data points) and then fine-tune the model on experimental data (20 870 data points). This training strategy enables SPT to accurately predict limiting activity coefficients even for unknown molecules, cutting the mean prediction error in half compared to state-of-the-art models for activity coefficient predictions such as COSMO-RS, UNIFAC, and improving on recent machine learning approaches.

Code available at: https://github.com/Bene94/SMILES2PropertiesTransformer; Data available at: https://polybox.ethz.ch/index.php/s/kyVOt3pwHW26PP4

References in corpus (3)

• Transfer learning for solvation free energies: from quantum chemistry to experiments
• SMILES Transformer: Pre-trained Molecular Fingerprint for Low Data Drug Discovery
• Machine Learning in Thermodynamics: Prediction of Activity Coefficients by Matrix Completion

Cited by in corpus (10)

• Graph Neural Networks for Temperature-Dependent Activity Coefficient Prediction of Solutes in Ionic Liquids
• SPT-NRTL: A physics-guided machine learning model to predict thermodynamically consistent activity coefficients
• Gibbs-Duhem-Informed Neural Networks for Binary Activity Coefficient Prediction
• HANNA: Hard-constraint Neural Network for Consistent Activity Coefficient Prediction
• Thermodynamics-Consistent Graph Neural Networks
• Predicting solvation free energies for neutral molecules in any solvent with openCOSMO-RS
• Gibbs-Helmholtz Graph Neural Network: capturing the temperature dependency of activity coefficients at infinite dilution
• GraphXForm: Graph transformer for computer-aided molecular design
• Prediction of Activity Coefficients by Similarity-Based Imputation using Quantum-Chemical Descriptors
• MLPROP -- an open interactive web interface for thermophysical property prediction with machine learning