Predicting Polymers’ Glass Transition Temperature by a Chemical Language Processing Model

Chen, Guang; Tao, Lei; Li, Ying

doi:10.3390/polym13111898

Cited by 56 publications

(39 citation statements)

References 81 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We have implemented a 2D CNN model based on images and revealed its poor generalization ability for T g prediction in the recent study . Besides this, we have also implemented the RNN model that is purely linguistic-based using the SMILES notation of a repeat unit as input . We re-evaluate our previously trained models using the new 566 MD simulations and compare them here with other models.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Tao

Varshney

2021

J. Chem. Inf. Model.

Self Cite

106

128

View full text Add to dashboard Cite

In the field of polymer informatics, utilizing machine learning (ML) techniques to evaluate the glass transition temperature T g and other properties of polymers has attracted extensive attention. This data-centric approach is much more efficient and practical than the laborious experimental measurements when encountered a daunting number of polymer structures. Various ML models are demonstrated to perform well for T g prediction. Nevertheless, they are trained on different data sets, using different structure representations, and based on different feature engineering methods. Thus, the critical question arises on selecting a proper ML model to better handle the T g prediction with generalization ability. To provide a fair comparison of different ML techniques and examine the key factors that affect the model performance, we carry out a systematic benchmark study by compiling 79 different ML models and training them on a large and diverse data set. The three major components in setting up an ML model are structure representations, feature representations, and ML algorithms. In terms of polymer structure representation, we consider the polymer monomer, repeat unit, and oligomer with longer chain structure. Based on that feature, representation is calculated, including Morgan fingerprinting with or without substructure frequency, RDKit descriptors, molecular embedding, molecular graph, etc. Afterward, the obtained feature input is trained using different ML algorithms, such as deep neural networks, convolutional neural networks, random forest, support vector machine, LASSO regression, and Gaussian process regression. We evaluate the performance of these ML models using a holdout test set and an extra unlabeled data set from high-throughput molecular dynamics simulation. The ML model’s generalization ability on an unlabeled data set is especially focused, and the model’s sensitivity to topology and the molecular weight of polymers is also taken into consideration. This benchmark study provides not only a guideline for the T g prediction task but also a useful reference for other polymer informatics tasks.

show abstract

Section: Resultsmentioning

confidence: 99%

“…In our case, we treat the SMILES notation of polymers as the input sequence to the RNN model and then specify the T g as the output token to be predicted. Our RNN model is realized and introduced in our recent study …”

Section: Datasets Models and Methodsmentioning

confidence: 99%

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Tao

Varshney

2021

J. Chem. Inf. Model.

Self Cite

106

128

View full text Add to dashboard Cite

show abstract

“…The integration of CGMD/ML is able to speed up the prediction of polymer properties at the chain level. Researchers have adopted ML models for the prediction of polymer properties mostly based on their monomer representation ( Ramprasad and Kim, 2019 ; Sattari et al, 2021 ; Chen et al, 2021b ; Gracheva et al, 2021 ), ignoring the influence of polymer chains, such as molecular weight, topology ( Tao et al, 2021a ), and copolymer sequence ( Kuenneth et al, 2021 ). Particularly for novel polymeric materials, there are limitations in the existing database due to unexplored chemical space ( Wilbraham et al, 2019 ).…”

Section: Application Of ML For Understanding and Design Of Polymer Chainsmentioning

confidence: 99%

Integration of Machine Learning and Coarse-Grained Molecular Simulations for Polymer Materials: Physical Understandings and Molecular Design

2022

Self Cite

View full text Add to dashboard Cite

In recent years, the synthesis of monomer sequence-defined polymers has expanded into broad-spectrum applications in biomedical, chemical, and materials science fields. Pursuing the characterization and inverse design of these polymer systems requires our fundamental understanding not only at the individual monomer level, but also considering the chain scales, such as polymer configuration, self-assembly, and phase separation. However, our accessibility to this field is still rudimentary due to the limitations of traditional design approaches, the complexity of chemical space along with the burdened cost and time issues that prevent us from unveiling the underlying monomer sequence-structure-property relationships. Fortunately, thanks to the recent advancements in molecular dynamics simulations and machine learning (ML) algorithms, the bottlenecks in the tasks of establishing the structure-function correlation of the polymer chains can be overcome. In this review, we will discuss the applications of the integration between ML techniques and coarse-grained molecular dynamics (CGMD) simulations to solve the current issues in polymer science at the chain level. In particular, we focus on the case studies in three important topics—polymeric configuration characterization, feed-forward property prediction, and inverse design—in which CGMD simulations are leveraged to generate training datasets to develop ML-based surrogate models for specific polymer systems and designs. By doing so, this computational hybridization allows us to well establish the monomer sequence-functional behavior relationship of the polymers as well as guide us toward the best polymer chain candidates for the inverse design in undiscovered chemical space with reasonable computational cost and time. Even though there are still limitations and challenges ahead in this field, we finally conclude that this CGMD/ML integration is very promising, not only in the attempt of bridging the monomeric and macroscopic characterizations of polymer materials, but also enabling further tailored designs for sequence-specific polymers with superior properties in many practical applications.

show abstract

“…Compared to the fully connected layers, the number of hidden neurons is drastically reduced by conventional and pool layers, thus allowing for far deeper networks. NNs have been applied to build ML models for the suitable selection of polymer-solvent pairs [99], glass transmission temperature [100] and thermal conductivity [25,77,95]. For the suitable selection of polymer-solvent pairs, a total of 11,958 polymer + good-solvent pairs and 8469 polymer + nonsolvent pairs were employed to train a binary classification NN model to judge whether a solvent is good or insoluble for a specific polymer [99].…”

Section: Algorithmmentioning

confidence: 99%

Review of machine learning‐driven design of polymer‐based dielectrics

Zhu

Deng

Lei

et al. 2021

IET Nanodielectrics

View full text Add to dashboard Cite

Polymer-based dielectrics are extensively applied in various electrical and electronic devices such as capacitors, power transmission cables and microchips, in which a variety of distinct performances such as the dielectric and thermal properties are desired. To fulfil these properties, the emerging machine learning (ML) technique has been used to establish a surrogate model for the structure-property linkage analysis, which provides an effective tool for the rational design of the chemical and morphological structure of polymers/nanocomposites. In this article, the authors reviewed the recent progress in the ML algorithms and their applications in the rational design of polymer-based dielectrics. The main routes for collecting training data including online libraries, experiments and high-throughput computations are first summarized. The fingerprints charactering the microstructures of polymers/nanocomposites are presented, followed by the illustration of ML models to establish a mapping between the fingerprinted input and the target properties. Further, inverse design methods such as evolution searching strategies and generative models are described, which are exploited to accelerate the discovery of new polymer-based dielectrics. Moreover, structure-property linkage analysis techniques such as Pearson correlation calculation, decision-tree-based methods and interpretable neural networks are summarized to identify the key features affecting the target properties. The future development prospects of the ML-driven design method for polymer-based dielectrics are also presented in this review. K E Y W O R D S fingerprinting, inverse design, machine learning, polymer-based dielectrics, structure-property linkage analysisThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

show abstract

Predicting Polymers’ Glass Transition Temperature by a Chemical Language Processing Model

Cited by 56 publications

References 81 publications

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature

Integration of Machine Learning and Coarse-Grained Molecular Simulations for Polymer Materials: Physical Understandings and Molecular Design

Review of machine learning‐driven design of polymer‐based dielectrics

Contact Info

Product

Resources

About