SolvBERT for solvation free energy and solubility prediction: a demonstration of an NLP model for predicting the properties of molecular complexes

SolvBERT для прогнозирования свободной энергии сольватации и растворимости: демонстрация NLP-модели для прогнозирования свойств молекулярных комплексов
Jiahui Yu, Chengwei Zhang, Yingying Cheng, Yun‐Fang Yang, Yuanbin She, Fengfan Liu, Weike Su, An Su
2023-01-01

SMILES representationSolvBERTTransformer modelssolubility predictionsolvation free energy
Deep learning models based on NLP, mainly the Transformer family, have been successfully applied to solve many chemistry-related problems, but their applications are mostly limited to chemical reactions. Meanwhile, solvation is an important concept in physical and organic chemistry, describing the interaction of solutes and solvents. This interaction leads to a solvation complex, a molecular complex similar to a reactant-reagent complex. In this study, we introduced the SolvBERT model, which reads the solute and solvents through the SMILES representation of the solvation complex. SolvBERT is pretrained in an unsupervised learning fashion using a large database of computational solvation free energies. The pretrained model can be used to predict the experimental solvation free energy or solubility, depending on the fine-tuning database. To the best of our knowledge, this multi-task prediction capability has not been observed in previously developed graph-based models for predicting the properties of molecular complexes. Furthermore, the performance of our SolvBERT in predicting solvation free energy is comparable to the state-of-the-art graph-based model DMPNN, mainly due to the clustering feature of the pretraining phase of the model, as demonstrated by the TMAP visualization algorithm.
1
Fine-tuning the pretrained model enables prediction of either experimental solvation free energy or solubility, demonstrating multi-task capability.
2
SolvBERT applies a Transformer-based NLP model to solvation complexes encoded as SMILES strings of solutes and solvents.
3
SolvBERT predicts solvation free energy with performance comparable to the state-of-the-art graph-based DMPNN model.
4
TMAP visualization indicates that clustering learned during pretraining contributes substantially to SolvBERT’s predictive performance.
5
The model is pretrained unsupervised on a large database of computational solvation free energies.

solute–solvent molecular complexes (solvation complexes)

prediction of solvation free energy and solubility

Publication Details
Publication Date
2023-01-01
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Jiahui Yu
Chengwei Zhang
Yingying Cheng
Yun‐Fang Yang
Yuanbin She
Fengfan Liu
Weike Su
An Su
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%