Generalizing property prediction of ionic liquids from limited labeled data: a one-stop framework empowered by transfer learning

Chen, Guzhong; Song, Zhen; Qi, Zhiwen; Sundmacher, Kai

doi:10.1039/d3dd00040k

Cited by 11 publications

(14 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Transfer learning in this manner enhances the model’s capacity to extrapolate valuable insights from limited, specialized data and demonstrates its potential for the advancement of molecular property prediction in the field of cheminformatics. Transfer learning is extensively examined in the field of GNNs. , …”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Graph Neural Networks and Structural Information on Ionic Liquids: A Cheminformatics Study on Molecular Physicochemical Property Prediction

Baran,

Kloskowski

2023

J. Phys. Chem. B

View full text Add to dashboard Cite

Ionic liquids (ILs) provide a promising solution in many industrial applications, such as solvents, absorbents, electrolytes, catalysts, lubricants, and many others. However, due to the enormous variety of their structures, uncovering or designing those with optimal attributes requires expensive and exhaustive simulations and experiments. For these reasons, searching for an efficient theoretical tool for finding the relationship between the IL structure and properties has been the subject of many research studies. Recently, special attention has been paid to machine learning tools, especially multilayer perceptron and convolutional neural networks, among many other algorithms in the field of artificial neural networks. For the latter, graph neural networks (GNNs) seem to be a powerful cheminformatic tool yet not well enough studied for dual molecular systems such as ILs. In this work, the usage of GNNs in structure–property studies is critically evaluated for predicting the density, viscosity, and surface tension of ILs. The problem of data availability and integrity is discussed to show how well GNNs deal with mislabeled chemical data. Providing more training data is proven to be more important than ensuring that they are immaculate. Great attention is paid to how GNNs process different ions to give graph transformations and electrostatic information. Clues on how GNNs should be applied to predict the properties of ILs are provided. Differences, especially regarding handling mislabeled data, favoring the use of GNNs over classical quantitative structure–property models are discussed.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Transfer learning is extensively examined in the field of GNNs. 28,29 Modeling and Neural Network Architecture. In this study, the architecture of the proposed network consisted of 4 graph convolution layers with 128, 256, 256, and 128 neurons followed by 2 linear (fully connected) layers.…”

Section: ■ Introductionmentioning

confidence: 99%

Graph Neural Networks and Structural Information on Ionic Liquids: A Cheminformatics Study on Molecular Physicochemical Property Prediction

Baran,

Kloskowski

2023

J. Phys. Chem. B

View full text Add to dashboard Cite

show abstract

“…The obtained Transformer-CNN model presents notably improved performance than the GC–COSMO approach. Very recently, a similar framework entitled ILTransR has been successfully developed for the prediction of 11 IL properties, which was pretrained on 10243410 IL-like SMILES and fine-tuned corresponding to each small labeled IL property data set (see Figure ).…”

Section: Structure–property Modelingmentioning

confidence: 99%

“…Very recently, this hierarchical ES screening framework was combined with that for ILs to present a large-scale comparative screening of the two types of neoteric solvents for CO 2 capture . Specifically, the toxicity estimation of ILs and salt component of ESs by the ILTransR model was added, followed by the experimental validation and the mechanism analysis based on quantum chemistry calculation. Through the comparative screening, it was indicated that ESs could be considered as a superior option to ILs for the physical absorption of CO 2 .…”

Section: Representative Applications Of Computer-aided Il Screening A...mentioning

confidence: 99%

“…On the basis of predicted property data and combined with quantum chemical calculations, potential IL candidates can be identified for potential applications (illustrated by cellulose dissolution, gas separation, extraction, and drug delivery). Very recently, Chen et al 128 proposed the ILTransR, a pretraining and finetuning model based on the Transformer-CNN architecture, which only requires the SMILES (and temperature and pressure if necessary) for IL property prediction. Through 11 benchmark case studies on different properties namely density, viscosity, melting point, glass transition temperature, thermal decomposition temperature, heat capacity, refractive index, surface tension, CO 2 solubility, cytotoxicity, and thermal conductivity, the superiority of the proposed framework was demonstrated.…”

Section: = +mentioning

confidence: 99%

See 1 more Smart Citation

Computer-Aided Molecular Design of Ionic Liquids as Advanced Process Media: A Review from Fundamentals to Applications

Song,

Chen,

Cheng

et al. 2023

Chem. Rev.

Self Cite

View full text Add to dashboard Cite

The unique physicochemical properties, flexible structural tunability, and giant chemical space of ionic liquids (ILs) provide them a great opportunity to match different target properties to work as advanced process media. The crux of the matter is how to efficiently and reliably tailor suitable ILs toward a specific application. In this regard, the computer-aided molecular design (CAMD) approach has been widely adapted to cover this family of high-profile chemicals, that is, to perform computer-aided IL design (CAILD). This review discusses the past developments that have contributed to the state-of-the-art of CAILD and provides a perspective about how future works could pursue the acceleration of the practical application of ILs. In a broad context of CAILD, key aspects related to the forward structure−property modeling and reverse molecular design of ILs are overviewed. For the former forward task, diverse IL molecular representations, modeling algorithms, as well as representative models on physical properties, thermodynamic properties, among others of ILs are introduced. For the latter reverse task, representative works formulating different molecular design scenarios are summarized. Beyond the substantial progress made, some future perspectives to move CAILD a step forward are finally provided.

show abstract

Enhancing aromatics extraction by double salt ionic liquids: Rational screening‐validation and mechanistic insights

Xie,

Chen,

Cheng

et al. 2023

AIChE Journal

Self Cite

View full text Add to dashboard Cite

Despite offering remarkable advantages as solvents, double salt ionic liquids (DSILs) have been scarcely studied for extractive dearomatization from hydrocarbons as well as many other applications, thus urging a theoretical guidance method. In this work, a systematic framework combining the rational screening validation and mechanistic analysis is proposed for tailoring DSILs for the o‐xylene/n‐octane separation. From an initial pool of commercially available ionic liquids (ILs), key thermodynamic properties of paired DSILs are predicted by COSMO‐RS while their important physical properties are estimated from those of corresponding parent ILs (retrieved from experimental database or predicted by a deep learning model). Promising DSILs are tested by liquid–liquid equilibrium experiments, wherein the effect of ion ratio is also evaluated. The mechanism underlying the tunability of DSIL thermodynamic properties is disclosed by quantum chemistry calculation and molecular dynamics simulation. This work can be a valuable reference for guiding the design of DSILs for diverse applications.

show abstract

Generalizing property prediction of ionic liquids from limited labeled data: a one-stop framework empowered by transfer learning

Abstract: Ionic liquids (ILs) could find use in almost every chemical process due to their wide spectrum of unique properties. The crux of the matter lies in whether a task-specific IL...

Cited by 11 publications

References 55 publications

Graph Neural Networks and Structural Information on Ionic Liquids: A Cheminformatics Study on Molecular Physicochemical Property Prediction

Graph Neural Networks and Structural Information on Ionic Liquids: A Cheminformatics Study on Molecular Physicochemical Property Prediction

Computer-Aided Molecular Design of Ionic Liquids as Advanced Process Media: A Review from Fundamentals to Applications

Enhancing aromatics extraction by double salt ionic liquids: Rational screening‐validation and mechanistic insights

Contact Info

Product

Resources

About