(362m) Hybrid, Interpretable Machine Learning for Thermodynamic Property Estimation Using Grammar2vec for Molecular Representation

Property prediction models have been developed for several decades with varying degrees of performance and complexity, from the group contribution-based methods to molecular simulations-based methods. An interesting issue in this area is finding an appropriate representation of molecules inherently suited for the property modeling problem. Here, we propose Grammar2vec, a SMILES grammar-based framework for generating dense, numeric molecular representations. Grammar2vec embeds molecular structural information contained in the grammar rules underlying SMILES string representations of molecules. We use Grammar2vec representations to build machine learning-based models for estimating boiling point (T_b) and critical temperature (T_c) and benchmark their performance against the popularly used group contribution (GC)-based methods. To ensure interpretability of the developed ML model, we perform a Shapley values-based analysis to estimate feature importance and simplify (or prune) the trained model.

References:

1. Venkatasubramanian, V. & Mann, V. Artificial intelligence in reaction prediction and chemical synthesis. Current Opinion in Chemical Engineering 36, 100749 (2022).

2. Gani, R. Group contribution-based property estimation methods: advances and perspectives. Current Opinion in Chemical Engineering 23, 184–196 (2019).

3. Mann, V. & Venkatasubramanian, V. Predicting chemical reaction outcomes: A grammar ontology-based transformer framework. AIChE Journal 67, e17190 (2021).

4. Mann, V. & Venkatasubramanian, V. Retrosynthesis prediction using grammar-based neural machine translation: An information-theoretic approach. Computers & Chemical Engineering 155, 107533. issn: 0098-1354 (2021).

5. Alshehri, A. S., Tula, A. K., You, F. & Gani, R. Next generation pure component property estimation models: With and without machine learning techniques. AIChE Journal, e17469 (2021).