Property Relationship Models Research Articles

Predicting the glass transition temperature of polymer based on generative adversarial networks and automated machine learning

AbstractSolution styrene‐butadiene rubber (SSBR) finds wide applications in high performance tire design and various other fields. This study aims to create a quantitative structure–property relationship (QSPR) model linking SSBR's glass transition temperature (Tg) to its structural properties. A dataset of 68 sets of data from published literature was compiled to develop a predictive machine learning model for SSBR's structural design and synthesis using small sample sizes. To tackle small sample sizes, a framework combining generative adversarial networks (GAN) and the Tree‐based Pipeline Optimization Tool (TPOT) is proposed. GAN is first used to generate additional samples that mirror the original dataset's distribution, expanding the dataset. The TPOT is then applied to automatically find the best model and parameter combinations, creating an optimal predictive model for the mixed dataset. Experimental results show that using GAN to enlarge the dataset and TPOT regression models significantly enhances model performance, increasing the R2 value from 0.745 to 0.985 and decreasing the RMSE from 7.676 to 1.569. The proposed GAN–TPOT framework demonstrates the potential of combining generative models with automated machine learning to improve materials science research. This combination accelerates research and development processes, enhances prediction and design accuracy, and introduces new perspectives and possibilities for the field.

Comprehensive accurate prediction of critical jet fuel properties with multiple machine learning models

Quantitative structure–property relationship (QSPR) model development driven by emerging machine learning (ML) shows promise for accelerating design and preparation of jet fuels with complex hydrocarbon compositions. In this work, we collected 104 jet fuels from different refineries, determined the detailed components of the fuel composition, and tested the fuel properties (density, viscosity, net heat of combustion, freezing point and flash point) using standard methods to form a database of molecule structure/composition and properties. Subsequently, six mainstream ML algorithms were adopted to establish the QSPR models, in which the prediction accuracy of the best ML models for each property is improved to above 0.93. Finally, the best ML property models are applied to predict unseen RP-3 fuels, and all prediction errors are within acceptable limits. This effort not only provides valuable data for the construction of the jet fuel database, but also provides tools for predicting its critical properties.

A GNN-Based QSPR Model for Surfactant Properties

Surfactants are among the most versatile molecules in the chemical industry because they can self-assemble in bulk solutions and at interfaces. Predicting the properties of surfactant solutions, such as their critical micelle concentration (CMC), limiting surface tension (γcmc), and maximal packing density (Γmax) at water–air interfaces, is essential to their rational design. However, the relationship between surfactant structure and these properties is complex and difficult to predict theoretically. Here, we develop a graph neural network (GNN)-based quantitative structure–property relationship (QSPR) model to predict the CMC, γcmc, and Γmax. Ninety-two surfactant data points, encompassing all types of surfactants—anionic, cationic, zwitterionic, and nonionic—are fed into the model, covering a temperature range of [20–30 °C], which contributes to its generalization across all surfactant types. We show that our models have high accuracy (R2 = 0.87 on average in tests) in predicting the three parameters across all types of surfactants. The effectiveness of the QSPR model in capturing the variation of CMC, γcmc, and Γmax with molecular design parameters are carefully assessed. The curated dataset, developed model, and critical assessment of the developed model will contribute to the development of improved surfactants QSPR models and facilitate their rational design for diverse applications.

Optimizing structure-property models of three general graphical indices for thermodynamic properties of benzenoid hydrocarbons

Cheminformatics is an interdisciplinary field that combines principles of chemistry, computer science, and information technology to process, store, analyze, and interpret chemical data. One area of cheminformatics is quantitative structure–property relationship (QSPR) modeling which is a computational approach that correlates the structural attributes of chemical compounds with their physical, chemical, or biological properties to predict the behavior and characteristics of new or untested compounds. Structure descriptors deliver contemporary mathematical tools required for QSPR modeling. One of a significant class of such descriptors is graph-based descriptors known as graphical descriptors. A degree-based graphical descriptor/invariant of a υ-vertex graph Ω=(VΩ,EΩ) has a general structure GDd=∑ij∈EΩπdegxi,degxj, where π is bivariate symmetric map, and degxi is the degree of vertex xi∈VΩ. For α∈R∖{0}, if π=(degxi×degxj)α (resp. π=(degxi+degxj)α, then GDd is called the general product-connectivity PCα (resp. sum-connectivity SCα) index of Ω. Moreover, the general Sombor index SOα has the structure π=(degxi2×degxj2)α. By choosing the heat capacity ΔH and the entropy E as representatives of thermodynamic properties, we in this paper find optimal value(s) of α which deliver the strongest potential of the predictors GDd∈{PCα,SCα,SOα} for predicting ΔH and E of benzenoid hydrocarbons. In order to achieve this, we employ tools such as discrete optimization and multivariate regression analysis. This, in turn, study completely solves two open problems proposed in the literature.

QSPRpred: a Flexible Open-Source Quantitative Structure-Property Relationship Modelling Tool

Building reliable and robust quantitative structure–property relationship (QSPR) models is a challenging task. First, the experimental data needs to be obtained, analyzed and curated. Second, the number of available methods is continuously growing and evaluating different algorithms and methodologies can be arduous. Finally, the last hurdle that researchers face is to ensure the reproducibility of their models and facilitate their transferability into practice. In this work, we introduce QSPRpred, a toolkit for analysis of bioactivity data sets and QSPR modelling, which attempts to address the aforementioned challenges. QSPRpred’s modular Python API enables users to intuitively describe different parts of a modelling workflow using a plethora of pre-implemented components, but also integrates customized implementations in a “plug-and-play” manner. QSPRpred data sets and models are directly serializable, which means they can be readily reproduced and put into operation after training as the models are saved with all required data pre-processing steps to make predictions on new compounds directly from SMILES strings. The general-purpose character of QSPRpred is also demonstrated by inclusion of support for multi-task and proteochemometric modelling. The package is extensively documented and comes with a large collection of tutorials to help new users. In this paper, we describe all of QSPRpred’s functionalities and also conduct a small benchmarking case study to illustrate how different components can be leveraged to compare a diverse set of models. QSPRpred is fully open-source and available at https://github.com/CDDLeiden/QSPRpred.Scientific ContributionQSPRpred aims to provide a complex, but comprehensive Python API to conduct all tasks encountered in QSPR modelling from data preparation and analysis to model creation and model deployment. In contrast to similar packages, QSPRpred offers a wider and more exhaustive range of capabilities and integrations with many popular packages that also go beyond QSPR modelling. A significant contribution of QSPRpred is also in its automated and highly standardized serialization scheme, which significantly improves reproducibility and transferability of models.

Machine Learning Approaches for Predicting Power Conversion Efficiency in Organic Solar Cells: A Comprehensive Review

Organic solar cells (OSCs), renowned for their lightweight, cost efficiency, and adaptability nature, stand out as a promising option for developing renewable energy. Improving the power conversion efficiency (PCE) of OSCs is essential, and researchers are delving into novel materials to achieve this. Traditional approaches are often laborious and costly, highlighting the need for predictive modeling. Machine learning (ML), especially via quantitative structure–property relationship (QSPR) models, is streamlining material development, with a goal to exceed a 20% PCE. In this review, the application of ML in OSCs is explored, and recent studies utilizing ML approaches for PCE prediction are reviewed, encompassing empirical functions, ML algorithms, self‐devised ML frameworks, and the combination with automated experimental technologies. First, the benefits of ML in predicting PCE for OSCs are addressed. Second, the development of high‐efficiency predictive models for both fullerene and nonfullerene acceptors is delved into. The impact of various ML algorithm models on PCE prediction is then assessed, taking into account the construction of predictive models. Moreover, the quality of databases and the selection of descriptors are considered. Databases and descriptors based on experimental studies are further categorized. Finally, prospects for the future development of OSCs are proposed.

From protein structure to an optimized chromatographic capture step using multiscale modeling.

Optimizing a biopharmaceutical chromatographic purification process is currently the greatest challenge during process development. A lack of process understanding calls for extensive experimental efforts in pursuit of an optimal process. In silico techniques, such as mechanistic or data driven modeling, enhance the understanding, allowing more cost-effective and time efficient process optimization. This work presents a modeling strategy integrating quantitative structure property relationship (QSPR) models and chromatographic mechanistic models (MM) to optimize a cation exchange (CEX) capture step, limiting experiments. In QSPR, structural characteristics obtained from the protein structure are used to describe physicochemical behavior. This QSPR information can be applied in MM to predict the chromatogram and optimize the entire process. To validate this approach, retention profiles of six proteins were determined experimentally from mixtures, at different pH (3.5, 4.3, 5.0, and 7.0). Four proteins at different pH's were used to train QSPR models predicting the retention volumes and characteristic charge, subsequently the equilibrium constant was determined. For an unseen protein knowing only the protein structure, the retention peak difference between the modeled and experimental peaks was 0.2% relative to the gradient length (60 column volume). Next, the CEX capture step was optimized, demonstrating a consistent result in both the experimental and QSPR-based methods. The impact of model parameter confidence on the final optimization revealed two viable process conditions, one of which is similar to the optimization achieved using experimentally obtained parameters. The multiscale modeling approach reduces the required experimental effort by identification of initial process conditions, which can be optimized.

Prediction of Dielectric Constant in Series of Polymers by Quantitative Structure-Property Relationship (QSPR)

This work is devoted to the investigation of dielectric permittivity which is influenced by electronic, ionic, and dipolar polarization mechanisms, contributing to the material’s capacity to store electrical energy. In this study, an extended dataset of 86 polymers was analyzed, and two quantitative structure–property relationship (QSPR) models were developed to predict dielectric permittivity. From an initial set of 1273 descriptors, the most relevant ones were selected using a genetic algorithm, and machine learning models were built using the Gradient Boosting Regressor (GBR). In contrast to Multiple Linear Regression (MLR)- and Partial Least Squares (PLS)-based models, the gradient boosting models excel in handling nonlinear relationships and multicollinearity, iteratively optimizing decision trees to improve accuracy without overfitting. The developed GBR models showed high R2 coefficients of 0.938 and 0.822, for the training and test sets, respectively. An Accumulated Local Effect (ALE) technique was applied to assess the relationship between the selected descriptors—eight for the GB_A model and six for the GB_B model, and their impact on target property. ALE analysis revealed that descriptors such as TDB09m had a strong positive effect on permittivity, while MLOGP2 showed a negative effect. These results highlight the effectiveness of the GBR approach in predicting the dielectric properties of polymers, offering improved accuracy and interpretability.

Innovative molecular descriptors in QSPR modeling: Integrating Carnahan-Starling EoS for predicting diffusion coefficients in hydrocarbons and mixtures

The rapid and accurate prediction of self-diffusion coefficients for pure fluids and mutual diffusion coefficients for binary mixtures remains a focal point of current research. In this work, molecular descriptors are innovatively defined by introducing molecular repulsive and attractive pressures based on the Carnahan-Starling equation of state. This approach aims to improve the quantitative structure–property relationship (QSPR) models for predicting thermophysical properties that vary with thermodynamic conditions. The shapely additive explanation method is employed for a comprehensive analysis of the model mechanism. Among the six molecular descriptors selected, our defined attractive pressure (patt), MATS2e and SpMin3_Bh(v) contribute significantly to the self-diffusion coefficient QSPR model. Based on this approach, QSPR models are constructed for 15 hydrocarbons and five of their binary mixtures. The results demonstrate that the GA-BPNN-based QSPR model for pure fluids achieves high prediction accuracy, with an external Rext2 of 0.978 and an external RMSEext of 5.349 × 10−9·m2·s−1. The optimal mixture descriptor type is x1d1+x2d2, and the GS-SVM-based mixture QSPR model predicts 98.24 % of values within a 5 % error margin, with Rext2 of 0.991 and RMSEext of 0.042 × 10−9·m2·s−1.

Synthesis and quantitative structure–activity relationship of unsaturated fatty acid amide propyl dimethyl tertiary amines and their application in clean fracturing fluid thickener in low temperature environment

Solve the problem of freezing and blocking of clean fracturing fluid thickener during winter fracturing operations in the north, which is conducive to improving the efficiency of fracturing operations and oil and gas production rate. In this paper, four kinds of fatty acid amidopropyl dimethyl tertiary amine were successfully prepared and characterized with FT-IR and 1H NMR. Their acid values and Krafft temperature are discussed. The molecular structures of the four kinds of fatty acid amidopropyl dimethyl tertiary amine salts were optimized using Gaussian09 software, then their single point energies were calculated. A quantitative structure–property relationship (QSPR) model was established to predict Krafft temperature of the four kinds of fatty acid amidopropyl dimethyl tertiary amine salts. N-(3-(Dimethylamino)propyl)oleamide (PKO-O), (9Z,12Z)-N-(3-(dimethylamino)propyl)octadeca-9,12-dienamide (PKO-LO) were selected as the main agent. The optimal thickener formulation O1-8 and O2-1 were optimized by single factor experiment and orthogonal test respectively. O2-1 thickener has high low-temperature stability at −5 °C. The viscosity increase property, gel breaking property, rheological property, and viscoelasticity of the clean fracturing fluid obtained by cross-linking O1-8 with O2-1 thickener were evaluated and the properties of O1-8 thickener are better. The result shows that the O1-8 thickener basically meets the conditions of the winter fracturing construction site, and has more practical application significance.

Heteroscedastic Gaussian Process Regression for material structure–property relationship modeling

Uncertainty quantification is a critical aspect of machine learning models for material property predictions. Gaussian Process Regression (GPR) is a popular technique for capturing uncertainties, but most existing models assume homoscedastic aleatoric uncertainty (noise), which may not adequately represent the heteroscedastic behavior observed in real-world datasets. Heteroscedasticity arises from various factors, such as measurement errors and inherent variability in material properties. Ignoring heteroscedasticity can lead to lower model performance, biased uncertainty estimates, and inaccurate predictions. Existing Heteroscedastic Gaussian Process Regression (HGPR) models often employ complicated structures to capture input-dependent noise but may lack interpretability. In this paper, we propose an HGPR approach that combines GPR with polynomial regression-based noise modeling to capture and quantify uncertainties in material property predictions while providing interpretable noise models. We demonstrate the effectiveness of our approach on both synthetic and physics-based simulation datasets, including mechanical properties (effective stress) of porous materials. We also introduce an approximated expected log predictive density method for model selection, which eliminates the need for retraining the model during leave-one-out cross-validation, allowing for efficient hyperparameter tuning and model evaluation. By capturing heteroscedastic behavior, enhancing interpretability, and improving model selection, our approach contributes to the development of more robust and reliable machine learning models for material property predictions, enabling informed decision-making in material design and optimization.

Prediction of CO2 solubility in aqueous amine solutions using machine learning method

In this study, a quantitative structure–property relationship (QSPR) model has been designed based on machine learning (ML) to offer a new method to accurately predict carbon dioxide (CO2) solubility in aqueous amine solutions. Molecular descriptors are used to denote representative features of the amine molecular structure supplemented with amine solution concentration, CO2 partial pressure, and temperature as inputs to the model. The coefficient of determination (R2) of the well-trained ML model reaches 0.971, and the average absolute deviation (AAD) of independent experimental validation is 4.785 %, effectively demonstrating the model’s reliability and generalization performance. Finally, SHapley Additive exPlanations (SHAP) is adopted to reveal the contribution of different features to the model predictions, making the model more transparent and interpretable. Overall work provides a novel, low-cost, efficient method to predict equilibrium CO2 solubility in aqueous amine solution and offers a new perspective in developing advanced amine for CO2 capture.

A Quantitative Structure–Property Relationship Model for Surface Tension Based on Artificial Neural Network

A Quantitative Structure–Property Relationship Model for Surface Tension Based on Artificial Neural Network

Descriptors-based machine-learning prediction of cetane number using quantitative structure–property relationship

The physicochemical properties of liquid alternative fuels are important but difficult to measure/predict, especially when complex surrogate fuels are concerned. In the present work, machine learning is used to develop quantitative structure–property relationship models. The fuel chemical structure is represented by molecular descriptors, allowing the linking of important features of the fuel composition and key properties of fuel utilization. Feature selection is employed to select the most relevant features that describe the chemical structure of the fuel and several machine learning algorithms are tested to construct interpretable models. The effectiveness of the methodology is demonstrated through the development of accurate and interpretable predictive models for cetane numbers, with a focus on understanding the link between molecular structure and fuel properties. In this context, matrix-based descriptors and descriptors related to the number of atoms in the molecule are directly linked with the cetane number of hydrocarbons. Furthermore, the results showed that molecular connectivity indices play a role in the cetane number for aromatic molecules. Also, the methodology is extended to predict the cetane number of ester and ether molecules, leveraging the design of alternative fuels towards fully sustainable fuel utilization.

Quantitative structure–property relationship (QSPR) modeling for evaluating fluorescence attributes across various aromatic heterocyclic compounds with ve-degree-based Sombor indices

Quantitative structure–property relationship (QSPR) modeling for evaluating fluorescence attributes across various aromatic heterocyclic compounds with ve-degree-based Sombor indices

Predicting drug solubility in organic solvents mixtures: A machine-learning approach supported by high-throughput experimentation

A novel approach based on supervised machine-learning is proposed to predict the solubility of drugs and drug-like molecules in mixtures of organic solvents. Similar to quantitative structure–property relationship (QSPR) models, different solvent types are identified by molecular descriptors, which, in this study, are considered as UNIFAC subgroups. To overcome the potential lack of UNIFAC subgroups for the complex Active Pharmaceutical Ingredients (APIs) currently developed in the pharmaceutical industry, the API molecule is considered as a unique entity in the proposed modelling approach. Therefore, API solubility is predicted as a function of temperature, functional subgroups of the solvents and composition of the solvent mixture; in turn, regressors’ correlation is handled through Partial Least-Squares (PLS) regression. The method is developed and tested with experimental data of a real API and 14 organic solvents that are industrially employed for crystallisation. Solubility predictions are accurate and precise for single solvents, binary mixtures and ternary mixtures of organic solvents at different compositions and temperatures, with a determination coefficient R2 ≥ 0.90.To further test the applicability of the model, the proposed approach is applied to 9 literature organic solubility datasets of drugs and drug-like compounds and compared to benchmark solubility models in the literature. Results show that the proposed approach provides satisfactory predictions: the majority of validation and calibration data have R2 = 0.95–0.99; the ratio between RMSE (root mean squared error) of the proposed method and the range of measured solubility values is from 1 to 3 orders of magnitude smaller than the RMSE ratio obtained by the benchmark models.

Graph neural networks for surfactant multi-property prediction

Surfactants are of high importance in different industrial sectors such as cosmetics, detergents, oil recovery and drug delivery systems. Therefore, many quantitative structure–property relationship (QSPR) models have been developed for surfactants. Each predictive model typically focuses on one surfactant class, mostly nonionics. Graph Neural Networks (GNNs) have exhibited a great predictive performance for property prediction of ionic liquids, polymers and drugs in general. Specifically for surfactants, GNNs can successfully predict critical micelle concentration (CMC), a key surfactant property associated with micellization. A key factor in the predictive ability of QSPR and GNN models is the data available for training. Based on extensive literature search, we create the largest available CMC database with 429 molecules and the first large data collection for surface excess concentration (Γm), another surfactant property associated with foaming, with 164 molecules. Then, we develop GNN models to predict the CMC and Γm and we explore different learning approaches, i.e., single- and multi-task learning, as well as different training strategies, namely ensemble and transfer learning. We find that a multi-task GNN with ensemble learning trained on all Γm and CMC data performs best. Thus, our results show that the simultaneous use of data from highly correlated properties can improve the predictability of surfactant properties for which only a small amount of experimental data is available. Finally, we test the ability of our CMC model to generalize on industrial grade pure component surfactants. The GNN yields highly accurate predictions for CMC, showing great potential for future industrial applications.

Evaluation of Various Approaches to Estimate Transplacental Clearance of Vancomycin for Predicting Fetal Concentrations using a Maternal-Fetal Physiologically Based Pharmacokinetic Model.

Evaluating drug transplacental clearance is vital for forecasting fetal drug exposure. Ex vivo human placenta perfusion experiments are the most suitable approach for this assessment. Various in silico methods are also proposed. This study aims to compare these prediction methods for drug transplacental clearance, focusing on the large molecular weight drug vancomycin (1449.3g/mol), using maternal-fetal physiologically based pharmacokinetic (m-f PBPK) modeling. Ex vivo human placenta perfusion experiments, in silico approaches using intestinal permeability as a substitute (quantitative structure property relationship (QSPR) model and Caco-2 permeability in vitro-in vivo correlation model) and midazolam calibration model with Caco-2 scaling were assessed for determining the transplacental clearance (CLPD) of vancomycin. The m-f PBPK model was developed stepwise using Simcyp, incorporating the determined CLPD values as a crucial input parameter for transplacental kinetics. The developed PBPK model of vancomycin for non-pregnant adults demonstrated excellent predictive performance. By incorporating the CLPD parameterization derived from ex vivo human placenta perfusion experiments, the extrapolated m-f PBPK model consistently predicted maternal and fetal concentrations of vancomycin across diverse doses and distinct gestational ages. However, when the CLPD parameter was derived from alternative prediction methods, none of the extrapolated maternal-fetal PBPK models produced fetal predictions in line with the observed data. Our study showcased that combination of ex vivo human placenta perfusion experiments and m-f PBPK model has the capability to predict fetal exposure for the large molecular weight drug vancomycin, whereas other in silico approaches failed to achieve the same level of accuracy.

Solubility evaluation of palm-based Mono-diacylglycerols (MDAGs) in food grade solvent (hexane, ethanol, acetone, water) using QSPR model approach

Mono-diacylglycerols (MDAGs) are emulsifiers widely used in the food industry. MDAGs based on refined bleached deodorized palm stearin (RBDPS) have not been widely developed. MDAGs have been synthesized using glicerolysis between RBDPS and glycerols. Due to their low solubility in food-grade solvents (hexane, ethanol, acetone, and water), it has been difficult to purify or fractionate MDAGs only by a simple method. This research attempted to develop a model of MDAGs solubility in food-grade solvents and understand the parameters that influence the solubility of MDAGs in food-grade solvents. The process used a combination of experimental data collection and quantitative structure–property relationship (QSPR) models. The QSPR model for MDAGs solubility in a single and binary solvent demonstrated that the descriptors of Δ dipole moment (Δµ), E HOMO, E LUMO, electronic energy, gap energy HOMO-LUMO (ΔE1, ΔE2) softness (S), electrophilicity index (ω), and electron transfer (ΔN) have a significant impact in determining the solubility of MDAGs. Both the single and the binary solvents solubility models can give satisfactory prediction results: the square of the correlation coefficient (R2pred) was 0.933 and 0.948, and the root-mean-square error (RMSE) was 0.075 and 0.031, respectively, for the whole data set. In addition, this paper provided a new and effective method for predicting the solubility of MDAGs in single and binary food-grade solvent systems.

Exploring the relationships between physiochemical properties of nanoparticles and cell damage to combat cancer growth using simple periodic table-based descriptors.

A comprehensive knowledge of the physical and chemical properties of nanomaterials (NMs) is necessary to design them effectively for regulated use. Although NMs are utilized in therapeutics, their cytotoxicity has attracted great attention. Nanoscale quantitative structure-property relationship (nano-QSPR) models can help in understanding the relationship between NMs and the biological environment and provide new ways for modeling the structural properties and bio-toxic effects of NMs. The goal of the study is to construct fully validated property-based models to extract relevant features for estimating and influencing the zeta potential and obtaining the toxicity profile regarding cell damage in the treatment of cancer cells. To achieve this, QSPR modeling was first performed with 18 metal oxide (MeOx) NMs to measure their materials properties using periodic table-based descriptors. The features obtained were later applied for zeta potential calculation (imputation for sparse data) for MeOx NMs that lack such information. To further clarify the influence of the zeta potential on cell damage, a QSPR model was developed with 132 MeOx NMs to understand the possible mechanisms of cell damage. The results showed that zeta potential, along with seven other descriptors, had the potential to influence oxidative damage through free radical accumulation, which could lead to changes in the survival rate of cancerous cells. The developed QSPR and quantitative structure-activity relationship models also give hints regarding safer design and toxicity assessment of MeOx NMs.