Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks
ACS Applied Materials & Interfaces · 2023-11-22 · 15 (46) · 54006-54017
DOI: 10.1021/acsami.3c13698 ↗At a glance
Journal article in ACS Applied Materials & Interfaces, published online on 2023-11-07. Written by Junyang Hu and 3 co-authors. Institution: Shanghai Key Laboratory of Advanced Polymeric Materials, School of Materials Science and Engineering, East China University of Science and Technology…. Cited 38 times so far (Crossref); 79 references. Funded by: National Natural Science Foundation of China.
Abstract (original language)
Establishing the structure-property relationship by machine learning (ML) models is extremely valuable for accelerating the molecular design of polymers. However, existing ML models for the polymers are subject to scarcity issues of training data and fewer variations of graph structures of molecules. In addition, limited works have explored the interpretability of ML models to infer the latent knowledge in the field of polymer science that could inspire ML-assisted molecular design. In this contribution, we integrate graph convolutional neural networks (GCNs) with data augmentation strategy to predict the glass transition temperature Tg of polymers. It is demonstrated that the data-augmented GCN model outperforms the conventional models and achieves a higher accuracy for the prediction of Tg despite a small amount of training data. Furthermore, taking advantage of molecular graph representations, the data-augmented GCN model has the capability to infer the importance of atoms or substructures from the understanding of Tg, which generally agrees with the experimental findings in the field of polymer science. The inferred knowledge of the GCN model is used to advise on the design of functional polymers with specific Tg. The data-augmented GCN model possesses prominent superiorities in the establishment of structure-property relationship and also provides an efficient way for accelerating the rational design of polymer molecules.
Subjects
Author keywords
Authors and institutions
- Shanghai Key Laboratory of Advanced Polymeric Materials, School of Materials Science and Engineering, East China University of Science and Technology, Shanghai 200237, China.
ORCID profiles: Jiaping Lin, Liangshun Zhang
Funding
- National Natural Science Foundation of China
Cite
APA
Hu, J., Li, Z., Lin, J., & Zhang, L. (2023). Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks. ACS Applied Materials & Interfaces, 15(46), 54006-54017. https://doi.org/10.1021/acsami.3c13698
Vancouver
Hu J, Li Z, Lin J, Zhang L. Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks. ACS Applied Materials & Interfaces. 2023;15(46):54006-54017. doi:10.1021/acsami.3c13698
BibTeX
@article{Hu2023,
author = {Hu, Junyang and Li, Zean and Lin, Jiaping and Zhang, Liangshun},
title = {Prediction and Interpretability of Glass Transition Temperature of Homopolymers by Data-Augmented Graph Convolutional Neural Networks},
journal = {ACS Applied Materials & Interfaces},
year = {2023},
volume = {15},
number = {46},
pages = {54006--54017},
doi = {10.1021/acsami.3c13698}
}
EUVIMED