Search

This paper reviews the application of machine learning to the inhibition of corrosion by organic molecules. The methodologies considered include quantitative structure‐property relationships (QSPR) and related data‐driven approaches. The characteristic features of their key components are considered as applied to corrosion inhibition, including datasets, response properties, molecular descriptors, machine learning methods, and structure‐property models. It is shown that the most important factors determining their choice and application features are: (1) the small or very small size of datasets, (2) the mechanism of corrosion inhibition associated with the adsorption of inhibitor molecules on the metal surface, and (3) multifactorial conditioning and noisiness of response property. On this basis, the application of machine learning to the inhibition of corrosion of materials based on iron, aluminum, and magnesium is considered. The main trends in the development of QSPR and related data‐driven modeling of corrosion inhibition are discussed, the shortcomings and common errors are considered, and the prospects for their further development are outlined. [ABSTRACT FROM AUTHOR]

BACKGROUND: Veterinary antibiotics are chemical compounds used to kill or inhibit the growth of pathogenic bacteria associated with animal diseases. These molecules can be defined by their retention times (tR) in liquid chromatography–mass spectrometry (LC–MS). One strategy to predict the tR of new veterinary antibiotics is the development of predictive quantitative structure–property relationships (QSPRs), which were used in this study. RESULTS: A database of 122 antibiotics was selected in which the tR was measured using a Hypersil GOLD column. An optimal three‐feature model was developed by integrating the unsupervised variable reduction, replacement method variable subset selection, and multiple linear regression. The negligible differences among the coefficient of determination and the root‐mean‐square error for the training set (R2 = 0.902 and RMSEC = 0.871) and test set (Q2 = 0.854 and RMSEP = 1.064) indicate a stable and predictive model. In a further step, a more in‐depth explanation of the mechanism of action of each descriptor in predicting the tR is provided, with the construction of the theoretical chemical space for accurate predictions of new antibiotics. CONCLUSION: The in silico model developed in this work identified three molecular descriptors associated with aqueous solubility, octanol–water partition coefficient, and the presence of negative and lipophilic atom pairs. The QSPR developed here could be implemented by agricultural and food chemists to identify and monitor existing and new antibiotics within the framework of LC–MS. The computational model was developed in accordance with five principles outlined by the Organization for Economic Co‐operation and Development. © 2024 Society of Chemical Industry. [ABSTRACT FROM AUTHOR]

Ionic liquids (ILs) are considered unique and attractive types of solvents with great potential to capture carbon dioxide (CO2) and reduce its emissions into the atmosphere. On the other hand, carrying out experimental measurements of CO2 solubility for each new IL is time‐consuming and expensive. Whereas, the possible combinations of cations and anions are numerous. Therefore, the preparation and design of such processes requires simple and accurate models to predict the solubility of CO2 as a greenhouse gas. In the present study, two different models, namely: artificial neural network (ANN) and support vector machine optimized with dragonfly algorithm (DA‐SVM) were used in order to establish a quantitative structure–property relationship (QSPR) between the molecular structures of cations and anions and the CO2 solubility. More than 10 116 CO2 solubility data measured in various ionic liquids (ILs) at different temperatures and pressures were collected. 13 significant PaDEL descriptors (E2M, MATS8S, TDB6I, TDB1S, ATSC4V, MATS8M, ATSC7V, Gats2S, Gats5S, Gats5C, ATSC6V, DE, and Lobmax), temperature and pressure were considered as the model input data. For the test set data (2023 data point), the estimated mean absolute error (MAE) and R2 for the ANN model are of 0.0195 and 0.9828 and 0.0219 and 0.9745 for the DA‐SVM model. The results obtained showed that both models can reliably predict the solubility of CO2 in ILs with a slight superiority of the ANN model. Examination of sensitivity and outlier diagnosis examinations confirmed that the QSPR model optimized using the ANN algorithm is better suited to correlate and predict this property. [ABSTRACT FROM AUTHOR]

Statistical learning is employed to present a principled framework for the establishment of quantitative structure‐property relationships (QSPR). Property predictions of industrial polymers formed by multiple reagents and at varying molecular weights are focused. A theoretical description of QSPR as well as a rigorous mathematical method is developed for the assimilation of experimental data. Results show that these methods can perform exceptionally well at establishing QSPR for glass transition temperature and intrinsic viscosity of polyesters. [ABSTRACT FROM AUTHOR]

Topological indices (TIs) can be used to forecast molecules' biological activity, physicochemical features, and toxicity. The use of topological indices and regression analysis can help us better understand drug behavior and contribute to the development of tailored drugs. This study investigates the predictive potential and efficacy of K-Banhatti and Zagreb type degree-based topological indices in quantitative structure–property relationship (QSPR) analysis of a comprehensive set of medications used for diabetes type-I and type-II disease. The K-Banhatti and Zagreb type degree-based topological indices are computed for 14 anti-diabetes drug molecules using edge/vertex partitioning techniques. By leveraging these topological indices, QSPR regression models are developed to predict the physicochemical properties of the understudy drugs. The results show that the values of these topological indices are highly correlated with certain physicochemical properties of the antidiabetes drugs. Furthermore, the comparative analysis revealed that, for all the considered properties except enthalpy of vaporization, Zagreb type indices outperform K-Banhatti indices with high predictive ability. Hence, it can be concluded that the Zagreb type indices are the best alternatives to theoretically predict the properties of anti-diabetes drugs. This theoretical analysis can help chemists in their right choice of the topological indices to theoretically predict the properties of anti-diabetes drugs without going into laborious experimentation. [ABSTRACT FROM AUTHOR]

Kinetic aqueous or buffer solubility is important parameter measuring suitability of compounds for high throughput assays in early drug discovery while thermodynamic solubility is reserved for later stages of drug discovery and development. Kinetic solubility is also considered to have low inter‐laboratory reproducibility because of its sensitivity to protocol parameters [1]. Presumably, this is why little efforts have been put to build QSPR models for kinetic in comparison to thermodynamic aqueous solubility. Here, we investigate the reproducibility and modelability of kinetic solubility assays. We first analyzed the relationship between kinetic and thermodynamic solubility data, and then examined the consistency of data from different kinetic assays. In this contribution, we report differences between kinetic and thermodynamic solubility data that are consistent with those reported by others [1, 2] and good agreement between data from different kinetic solubility campaigns in contrast to general expectations. The latter is confirmed by achieving high performing QSPR models trained on merged kinetic solubility datasets. The poor performance of QSPR model trained on thermodynamic solubility when applied to kinetic solubility dataset reinforces the conclusion that kinetic and thermodynamic solubilities do not correlate: one cannot be used as an ersatz for the other. This encourages for building predictive models for kinetic solubility. The kinetic solubility QSPR model developed in this study is freely accessible through the Predictor web service of the Laboratory of Chemoinformatics (https://chematlas.chimie.unistra.fr/cgi‐bin/predictor2.cgi). [ABSTRACT FROM AUTHOR]

Pyridine and its derivatives have been applied clinically for the treatment of a wide range of diseases and in the synthesis of novel drugs. In the present work, imidazo[4,5‐b]pyridine derivatives as anticancer drugs were exhibited to select the important descriptor for quantum chemical properties. Two of the fundamental thermodynamic properties are heat capacity (Cv) and entropy (S), which are important in the field of chemical kinetics and are key in the understanding and design of chemical processes involving chemical reactions. A Quantitative Structure–Property Relationship (QSPR) study was used to predict the quantum chemical properties like Cv and S of 105 imidazole derivatives using molecular descriptors and the genetic algorithm–multiple linear regression (GA–MLR). The best QSPR models were selected using criteria coefficients such as R2, R2adj, RMSE and Fisher ratio. Different internal and external validation metrics were adopted to evaluate the stability, fit and predictive power of the QSPR models. The validation results and statistical analysis show that the models possess good prediction power and robustness, and the total size (TS) and Sanderson electronegativity(RDF060e) and total information content index(TIC1) of imidazo[4,5‐b]pyridine derivatives are increasingly related to the studied properties. [ABSTRACT FROM AUTHOR]

Aluminophosphate glasses have found wide applications in various fields, such as biomedical materials, optical components, sealing materials, and nuclear waste forms. In spite of their well‐investigated short‐range ordered structures, the relationship between the properties and the medium‐range structural features is far from being understood. In this paper, atomistic structures of sodium aluminophosphate (SAP) glasses were reproduced by molecular dynamics simulations. In addition, experimental methods, including Raman, differential scanning calorimetry, and synchrotron X‐ray total scattering, have been applied to characterize the structures of these glasses, together with the measurements of various glass properties, such as the density, glass transition temperature (Tg), coefficient of thermal expansion (CTE), and hardness. Moreover, the quantitative structure–property relationship (QSPR) analysis was performed to correlate the simulated glass structures with the experimentally measured properties. The simulation results reveal that the P–O–P linkages in the glass network are gradually replaced by the P–O–Al linkages with additional alumina to the compositions, which contributes to the property changes of the SAP glass systems. Meanwhile, the long chains in the SAP glasses tend to form ring structures, and the primitive rings are concentrated in the range between 4‐ and 20‐membered rings. Furthermore, QSPR analysis shows that the simulated structures have good correlations with the experimental properties, and the established structure–property model is promising in predicting certain properties of aluminophosphate glass systems. [ABSTRACT FROM AUTHOR]

Conjugated QSPR models for reactions integrate fundamental chemical laws expressed by mathematical equations with machine learning algorithms. Herein we present a methodology for building conjugated QSPR models integrated with the Arrhenius equation. Conjugated QSPR models were used to predict kinetic characteristics of cycloaddition reactions related by the Arrhenius equation: rate constant logk ${{\rm l}{\rm o}{\rm g}k}$ , pre‐exponential factor logA ${{\rm l}{\rm o}{\rm g}A}$ , and activation energy Ea ${{E}_{{\rm a}}}$. They were benchmarked against single‐task (individual and equation‐based models) and multi‐task models. In individual models, all characteristics were modeled separately, while in multi‐task models logk ${{\rm l}{\rm o}{\rm g}k}$ , logA ${{\rm l}{\rm o}{\rm g}A}$ and Ea ${{E}_{{\rm a}}}$ were treated cooperatively. An equation‐based model assessed logk ${{\rm l}{\rm o}{\rm g}k}$ using the Arrhenius equation and logA ${{\rm l}{\rm o}{\rm g}A}$ and Ea ${{E}_{{\rm a}}}$ values predicted by individual models. It has been demonstrated that the conjugated QSPR models can accurately predict the reaction rate constants at extreme temperatures, at which reaction rate constants hardly can be measured experimentally. Also, in the case of small training sets conjugated models are more robust than related single‐task approaches. [ABSTRACT FROM AUTHOR]

A topological index, also known as a connectedness index, is a numerical characteristic of chemical structure that is computed by using its molecular graph in theoretical chemistry. In quantitative structure‐activity relationships (QSARs), which relate a molecule's chemical structure to its physical and chemical properties, topological indices are employed. The graph entropies with topological indices were inspired by Shannon's entropy concept and became the information‐theoretic quantities for measuring the structural information of chemical graphs. The theory of graphs is useful in determining the relationship between specific properties of chemical structures using different graph entropy measures. Through these entropies, many physical and chemical characteristics, such as melting point, energy generation, Henry's Law, and molar mass of chemical compounds, can be calculated. For this, the quantitative structure‐property relationship (QSPR) models are designed using ve‐degree‐based entropies to examine some physical properties of benzene derivatives. For the computation of entropies, a Maple‐based program is developed. The QSPR study is performed using SPSS and linear regression techniques. In this work, we have observed that the redefined third Zagreb entropy, the Balaban entropy, and the Randic entropy are the best predictors. The physiochemical characteristics, namely Henry's Law and critical pressure, can be predicted by Randic entropy and Balaban entropy, respectively. The redefined third Zagreb entropy can predict four other properties: enthalpy, molar mass, π‐electronic energy, and molecular weight. [ABSTRACT FROM AUTHOR]

The present work was devoted to explore the quantitative structure‐property relationships for gas‐to‐ionic liquid partition coefficients (log KILA). A series of linear models were first established for the representative dataset (IL01). The optimal model was a four‐parameter equation (1Ed) consisting of two electrostatic potential‐based descriptors (ΣVs,ind- ${{\rm { \Sigma }}{V}_{s,ind}^{-}}$ and Vs,max), one 2D matrix‐based descriptor (J_D/Dt) and dipole moment (μ). All of the four descriptors introduced in the model can find the corresponding parameters, directly or indirectly, from Abraham's linear solvation energy relationship (LSER) or its theoretical alternatives, which endows the model good interpretability. Gaussian process was utilized to build the nonlinear model. Systematical validations, including 5‐fold cross‐validation for the training set, the validation for test set, as well as a more rigorous Monte Carlo cross‐validation were performed to verify the reliability of the constructed models. Applicability domain of the model was evaluated, and the Williams plot revealed that the model can be used to predict the log KILA values of structurally diverse solutes. The other 13 datasets were also processed in the same way, and all of the linear models with expressions similar to equation 1Ed were obtained. These models, whether linear of nonlinear, represent satisfactory statistical results, which confirms the universality of the method adopted in this study in QSPR modeling of gas‐to‐IL partition. [ABSTRACT FROM AUTHOR]

The quantitative structure–property relationship (QSPR) is a fundamental technique for evaluating and screening potentially valuable molecules in the field of drug discovery. There is an urgent need to speed up pharmaceutical research and development and a huge chemical space to explore, which necessitate effective and precise computer‐aided QSPR modeling methods. Previous studies with various deep learning models are limited because they are trained on separate small datasets, known as the small‐sample problem. Using transfer learning, this article describes a sparse sharing method that uses advanced graph‐based models to construct an efficient and reasonable multitask learning workflow for QSPR prediction. The proposed workflow is systematically and comprehensively tested with four benchmark datasets containing different targets, and several precisely predicted molecular examples are illustrated. The results demonstrate that an obvious improvement in the prediction of molecular properties is achieved, along with the ability to predict multiple properties simultaneously. [ABSTRACT FROM AUTHOR]

The combination of the generical molecular maps of atom‐level properties (MOLMAPs) encoding approach and the Random Forest algorithm (RF) is applied in order to model, predict, and interpret the structural motifs responsible for a certain organic molecule's melting point (mp) profile. A high‐quality database is used for model build‐up and evaluation of predictive ability. The obtained results for the complete independent test set (R2 = 0.811, MAE = 31.99 K, RMS = 43.98 K) are comparable or better than reference works. The form of codification represents implicitly the structure of a given molecule and highlights the interactions responsible for a certain melting point profile. This generical encoding approach groups different structural motifs based on its calculated atomic‐based properties leading to good predictive ability for structurally different chemical systems not contained in the training set. [ABSTRACT FROM AUTHOR]

Quantitative structure–property relationship (QSPR) modeling was investigated to predict drug and drug‐like compounds solubility in supercritical carbon dioxide. A dataset of 148 drug\drug‐like compounds, accounting for 3971 experimental data points (EDPs), was collected and used for modelling the relationship between selected molecular descriptors and solubility fraction data achieved by a nonlinear approach (Artificial neural network, ANN) based on molecular descriptors. Experimental solubility data for a given drug were published as a function of temperature and pressure. In the present study, 11 significant PaDEL descriptors (AATS3v, MATS2e, GATS4c, GATS3v, GATS4e, GATS3 s, nBondsM, AVP‐0, SHBd, MLogP, and MLFER_S), the temperature and the pressure were statistically proved to be sufficient inputs. The architecture of the optimized model was found to be {13,10,1}. Several statistical metrics, including average absolute relative deviation (AARD=3.7748 %), root mean square error (RMSE=0.5162), coefficient of correlation (r=0.9761), coefficient of determination (R2=0.9528), and robustise (Q2=0.9528) were used to validate the obtained model. The model was also subjected to an external test by using 143 EDPs. Sensitivity analysis and domain of application were examined. The overall results confirmed that the optimized ANN‐QSPR model is suitable for the correlation and prediction of this property. [ABSTRACT FROM AUTHOR]

Density functional theory calculations on saturated monobasic ketones, aldehydes, and ethers were carried out using Gaussian‐16 software. For each compound, the most stable molecule with the lowest energy and no imaginary frequency was found, and the molecular descriptors were calculated using the Alvadesc software based on the optimized geometry. The stepwise MLR method was used to develop a four‐parameter linear regression equation between the critical temperatures and molecular descriptors of 50 compounds in the training set, with a correlation coefficient of.9991. The QSPR model was robust and reliable, according to the Fisher and student‐t tests, the Durbin‐Watson test, cross‐validation and external validation, and application domain analysis. [ABSTRACT FROM AUTHOR]

There has been a remarkable increase in the number of biologics, especially monoclonal antibodies, in the market over the last decade. In addition to attaining the desired binding to their targets, a crucial aspect is the 'developability' of these drugs, which includes several desirable properties such as high solubility, low viscosity and aggregation, physico‐chemical stability, low immunogenicity and low poly‐specificity. The lack of any of these desirable properties can lead to significant hurdles in advancing them to the clinic and are often discovered only during late stages of drug development. Hence, in silico methods for early detection of these properties, particularly the ones that affect aggregation and solubility in the earlier stages can be highly beneficial. We have developed a computational framework based on a large and diverse set of protein specific descriptors that is ideal for making liability predictions using a QSPR (quantitative structure‐property relationship) approach. This set offers a high degree of feature diversity that may coarsely be classified based on (1) sequence (2) structure and (3) surface patches. We assess the sensitivity and applicability of these descriptors in four dedicated case studies that are believed to be representative of biophysical characterizations commonly employed during the development process of a biologics drug candidate. In addition to data sets obtained from public sources, we have validated the descriptors on novel experimental data sets in order to address antibody developability and to generate prospective predictions on Adnectins. The results show that the descriptors are well suited to assist in the improvement of protein properties of systems that exhibit poor solubility or aggregation. [ABSTRACT FROM AUTHOR]

Quantitative Structure‐Property Relationships (QSPRs) have found applications in many areas of chemistry and engineering as effective prediction methods. QSPRs use molecular descriptors to simplify complex molecular properties to a single value and have been used extensively for constant value properties. Liquid heat capacity (cpl) is another property where QSPRs can be helpful prediction tools. Researchers have shown strong correlation between the cpl and various molecular descriptors, but these predictions are limited to a single temperature, usually 298.15 K. Additionally, other QSPRs have had problems with oxygen‐containing functional groups. In this work, QSPRs for cpl at various temperatures were developed using data selected from the DIPPR database using a novel search method. This method improves on existing QSPRs for cpl by using unique descriptors but does not overcome the issue of oxygen‐containing species. [ABSTRACT FROM AUTHOR]

Summary: Photovoltaic technology attracts researchers from industry and academia due to its potential in producing electricity directly from the sunlight. Among all the photovoltaic devices, the dye‐sensitized solar cell has gained preference due to its low‐cost fabrication and versatility in electrolytes, dye, substrate, and catalyst. The optical, electrical, and structural properties of the materials determine the power conversion efficiency of a solar cell. But, conducting experiments in the laboratory to identify the suitable materials for the fabrication of an efficient solar cell requires much time, cost, and human effort. The proven potential of machine learning techniques in pattern matching and computer vision motivated the researchers to employ these techniques for predicting the efficiency of solar cells. The research works conducted so far show the applications of these techniques in predicting the optimum efficiency, best suitable design, and material for the fabrication of Dye‐Sensitized Solar Cells (DSSCs). In this paper, the authors present a comprehensive review of the machine learning techniques employed and the types of input data used for predicting the design and efficiency of solar cells. They also give essential insights into the selection of optimum parameters for selecting the materials for fabricating a substrate, dye sensitizer, semiconductor, electrolyte, and catalyst for designing the most efficient dye‐sensitized solar cell without conducting experiments in the laboratory. This paper may prove a time and cost‐saving assistant for developing a customized neural network model for predicting the efficiency of a DSSC from the dataset available in the literature. Highlights: Artificial Neural Network (ANN) model is useful to identify the suitable materials for efficient DSSC assembly.The tailored neural network models minimize the need for hit and trial experiments.The hybrid of ANN and Genetic Algorithm (GA) offers a low‐cost technological solution for DSSC assembly.The experimental data are vital for extracting useful insights for DSSC fabrication. [ABSTRACT FROM AUTHOR]

Deep learning and graph‐based models have gained popularity in various life science applications such as property modeling, achieving state‐of‐the‐art performance. However, the quantification of prediction uncertainty in these models is less studied and is not applied in the low dataset size regime, which characterizes many chemical engineering‐related molecular properties. In this work, we apply two graph‐based models to model the critical‐ temperature, pressure, and volume and apply three techniques (the bootstrap, the ensemble, and the dropout) to quantify the prediction uncertainty. The overall model confidence is evaluated using the coverage. The results suggest that graph‐based models perform better compared with current group‐contribution‐based property modeling techniques while eliminating the tedious task of developing molecular descriptors. [ABSTRACT FROM AUTHOR]

Lipophilicity, as quantified by the decimal logarithm of the octanol–water partition coefficient (log KOW), is an essential environmental property. Deep neural networks (DNNs) based quantitative structure–property relationship (QSPR) studies have received more and more attention because of their excellent performance for prediction. However, the black‐box nature of DNNs limits the application range where interpretability is essential. Hence, this study aims to develop an accurate and interpretable deep neural network (AI‐DNN) model for log KOW prediction. A hybrid method of molecular representation was employed to guarantee the accuracy of the proposed AI‐DNN model. The hybrid molecular representations are able to integrate the directed message passing neural networks (D‐MPNNs) learned molecular representations and the fixed molecule‐level features of CDK descriptors, and can capture both the local and the global features of overall molecule. The performance analysis shows that the proposed QSPR model exhibits promising predictive accuracy and discriminative power in the structural isomers and stereoisomers. Moreover, the Monte Carlo Tree Search (MCTS) approach was used to interpret the proposed AI‐DNN model by identifying the molecular substructures contributed to the lipophilicity. This interpretability can be applied to critical fields where there is a high demand for interpretable deep networks, such as green solvent design and drug discovery. [ABSTRACT FROM AUTHOR]

The quantitative structure‐property relationship (QSPR) approach has widely been used to predict several physicochemical properties of materials employing the information obtained from their chemical structures (numerical descriptors). In the present work, we have generated three individual QSPR models for three different endpoints for a large number of polymers in order to determine their fire retardant property such as heat release capacity, total heat release, and %Char, using the only two‐dimensional descriptors with definite physicochemical meaning. Relevant subsets of descriptors were selected employing a genetic algorithm approach; subsequently, the selected descriptors were utilised for the identification of the best combination of the variables for the model generation, while the final models were developed employing the partial least squares (PLS) regression algorithm. The generated models were rigorously validated using various internationally accepted internal and external validation metrics. All the models showed promising statistical quality in terms of determination coefficient R2 (0.802, 0.842 and 0.826), cross‐validated leave‐one‐out Q2 (0.759, 0.810 and 0.752) and predictive R2pred or Q2ext (0.810, 0.900 and 0.847) for HRC (nTraining=62, nTest=28), THR (nTraining=64, nTest=21) and %char (nTraining=49, nTest=21) datasets, respectively. All the certified models were used for prediction of flammability characteristics of 37 external set compounds, and further, the quality of prediction was determined by using the PRI software tool. The final models of HRC, THR and %Char formation of polymers may be useful to predict the flammability characteristics of polymers quickly before their synthesis and used as a better alternative approach to the experimental testing of flammability of polymers. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) based quantitative structure–property relationship (QSPR) studies are receiving increasing attention due to their excellent performances. A systematic methodology coupling multiple machine learning technologies is proposed to systematically solve vital problems including applicability domain and prediction uncertainty in DNN‐based QSPR modeling. Key features are rapidly extracted from plentiful but chaotic descriptors by principal component analysis (PCA) and kernel PCA. Then, a detailed applicability domain (AD) is defined by K‐means algorithm to avoid unreliable predictions and discover its potential impact on prediction uncertainty. Moreover, prediction uncertainty is analyzed with dropout‐embedded DNN by thousands of independent tests to assess the reliability of predictions. The prediction of flashpoint temperature is employed as a case study, demonstrating that the model accuracy is remarkably improved comparing with the referenced model. Furthermore, the proposed methodology breaks through difficulties in analyzing the uncertainty of DNN‐based QSPRs and presents an AD correlated with the uncertainty. [ABSTRACT FROM AUTHOR]

Cancer is a deadly disease. There is no drug yet for the treatment of some types of cancer. New drug trials are being conducted to treat this disease. One of them is HDAC‐based multi‐target drugs. Topological indices (TIs) are the numerical descriptors of a molecular structure obtained by the molecular graph. TIs can be used in the structure–property relationship (QSPR) and structure–activity relationship studies to provide information about the physicochemical properties and biological properties of molecules. In this paper, vorinostat, tucidinostat, triciferol, and CUDC‐101 and CUDC‐907, which are seen as an important light in cancer treatment, are studied. M‐polynomials based on degree and NM‐polynomials based on the sum of degrees of neighboring vertices for chemical graphs of these drugs are calculated. By using these polynomials, many TIs that depend on the degree and the sum of neighboring degrees are obtained. Moreover, The QSPR models are designed using these TIs to estimate some physicochemical properties of these drugs. The QSPR studies are obtained using the curvilinear regression method. [ABSTRACT FROM AUTHOR]

In‐silico screening of novel biofuel molecules based on chemical and fuel properties is a critical first step in the biofuel evaluation process due to the significant volumes of samples required for experimental testing, the destructive nature of engine tests, and the costs associated with bench‐scale synthesis of novel fuels. Predictive models are limited by training sets of few existing measurements, often containing similar classes of molecules that represent just a subset of the potential molecular fuel space. Software tools can be used to generate every possible molecular descriptor for use as input features, but most of these features are largely irrelevant and training models on datasets with higher dimensionality than size tends to yield poor predictive performance. Feature selection has been shown to improve machine learning models, but correlation‐based feature selection fails to provide scientific insight into the underlying mechanisms that determine structure–property relationships. The implementation of causal discovery in feature selection could potentially inform the biofuel design process while also improving model prediction accuracy and robustness to new data. In this study, we investigate the benefits causal‐based feature selection might have on both model performance and identification of key molecular substructures. We found that causal‐based feature selection performed on par with alternative filtration methods, and that a structural causal model provides valuable scientific insights into the relationships between molecular substructures and fuel properties. [ABSTRACT FROM AUTHOR]

Research on the solubility of n‐propanol in different solvents is important because of the wide applications of n‐propanol in cosmetics, cleaning, printing, coatings, and chemical industries. The goal of this research was to investigate the structural or empirical properties of different solvents on the Ostwald solubility coefficient of n‐propanol. In this research, two different approaches, that is, linear solvation energy relationship (LSER) and quantitative structure–property relationship (QSPR), have been used to predict and descript the gas‐solvent partition coefficient of n‐propanol, as its solubility index, in 36 different solvents. In the suggested LSER model, the role of solvent–solute interactions on the solubility of n‐propanol as a physicochemical property can be investigated. In addition to structural features, which were used in the QSPR model, the interpretation of the LSER model showed that the donor number of solvents, modified polarity scale of solvents, and entropy of solvation are three important parameters in the Ostwald solubility coefficient of n‐propanol. [ABSTRACT FROM AUTHOR]

Thermoelectric materials with a high Seebeck coefficient, high electrical conductivity, and low thermal conductivity are required to directly and efficiently convert unused heat into electricity. In this study, we construct models predicting the Seebeck coefficient, electrical conductivity, and thermal conductivity using existing material databases. In addition to the ratios of atoms in the crystals and temperature at which the materials are used, the values from the X‐ray diffraction (XRD) spectra were used as inputs to represent the crystal structure of the materials. It was confirmed that the constructed models could predict the properties with high accuracy using the X‐ray diffraction values. Additionally, using the constructed models, we succeeded in proposing promising new candidate materials with high Seebeck coefficients, high electric conductivities, and low thermal conductivities. [ABSTRACT FROM AUTHOR]

Support vector regression (SVR) is able to consider the nonlinear relationship between explanatory variables X and a target variable y to build a regression model with high predictive accuracy. Additionally, y values predicted with SVR models for new samples can exceed the actual y values in training data. However, because the Gaussian kernel, which is a kernel function generally used in SVR, is based on the Euclidean distance between samples, it is unable to consider the importance of X when building the regression model. Therefore, in this study, the focus was on the importance of X that can be calculated by random forests (RF), and a novel SVR method, called variable importance‐considering support vector regression (VI‐SVR), was proposed based on this importance. Because X is weighted based on importance, the greater the importance of X, the greater its contribution to the predicted value. Analysis using the spectral, quantitative structure–property relationship (QSPR), and quantitative structure–activity relationship (QSAR) datasets confirmed that the predictive accuracy of VI‐SVR was better than that of SVR. VI‐SVR Python code is available at https://github.com/hkaneko1985/dcekit. We proposed a novel SVR method, called variable importance‐considering support vector regression (VI‐SVR), based on this importance calculated with random forests. Because X is weighted based on importance, the greater the importance of X, the greater its contribution to the predicted value. Analysis using the spectral, QSPR, and QSAR datasets confirmed that the predictive accuracy of VI‐SVR was better than that of SVR. [ABSTRACT FROM AUTHOR]

The combined quantum mechanics/molecular mechanics (QM/MM) simulations of equilibrium geometry configurations followed by electron density analysis provide reliable quantitative structure‐property relationship equations to estimate the reactivity of compounds in the active sites of enzymes. The main drawback is high computational cost of such calculations. Here, we report on a benchmark study aiming to optimize computational protocol for the accuracy of predictions. We considered an important example of cephalosporin hydrolysis in the active site of L1 metallo‐β‐lactamase and found that it is important to consider contributions to the one‐electron part of the QM Hamiltonian from all MM system rather than using the cutoff of electrostatic interactions. Switching from the reference PBE0‐D3/6‐31G(d,p) QM protocol to the reduced PBE0‐D3/6‐31G scheme decreases the number of basis set functions by almost twice, increasing the error of the rate constant estimates up to 18 seconds−1 compared with the reference 10 seconds−1. Therefore, the QM(PBE0‐D3/6‐31G)/MM(AMBER) level of theory can be recommended for estimates of cephalosporin reactivity in the search of new antibiotics. [ABSTRACT FROM AUTHOR]

In the present investigation, quantitative structure–property relationship (QSPR) modeling was carried out on 48 aliphatic esters to develop a robust model for the prediction of thermodynamic properties such as the enthalpy of vaporization at standard condition (∆H°vap kJ mol−1) and normal temperature of boiling points (Tbp° K). Multiple linear regression (MLR) and backward (BW) stepwise regression methods were used to select the descriptors derived from the Chemicalize program to give the QSPR models. These models were used to delineate the important descriptors responsible for the properties of the aliphatic esters. The multicollinearity and autocorrelation properties of the descriptors used in the models were tested by calculating the variance inflation factor, Pearson correlation coefficient, and the Durbin–Watson statistics. Leave‐one‐out cross‐validation, leave‐group (fivefold)‐out, and external validation criteria (Q2F1, Q2F2, Q2F3, CCC, R2m) were proposed to verify the predictive performance of QSPR models derived by BW‐MLR analysis. The predictive ability of the models was found to be satisfactory. Thus, QSPR models derived from this study may be helpful for modeling and designing some new aliphatic esters and predicting their properties. [ABSTRACT FROM AUTHOR]

We establish a QSPR analysis for the bioconcentration factor of 851 heterogeneous structural compounds. Linear models are proposed via two different approaches: i. the optimal descriptor method implemented in CORAL, and ii. multivariable linear regressions on the best molecular descriptors found with the Replacement Method on 44,216 structural descriptors. Such variables are derived with different freely available softwares, such as PaDEL, PyDescriptor, Mold2, QuBiLs‐MAS and ISIDA/Fragmentor. The same validation set is employed in order to compare the predictive performance between the so‐obtained CORAL and RM based models. Finally, the comparison of several models for the bioconcentration factor confirms the ability of the so‐called index of ideality of correlation to be a criterion of predictive potential in Quantitative Structure‐Property Relationships. [ABSTRACT FROM AUTHOR]

Pyridines are compounds with a wide range of biological activities and they are the basis of several groups of drugs. In this study, the relationship between molecular descriptors and the thermal energy (Eth kJ/mol), heat capacity (Cv J/molK), entropy (S J/mol K), enthalpy of formation (ΔHf kJ/mol), and Gibbs free energy (ΔG kJ/mol) of pyridine derivatives is studied. These properties were calculated at the Hartree–Fock level of theory and 6‐311G* G basis sets by Gaussian 09 software. The chemical structures of 171 pyridine derivatives were drawn by the Gauss View 05 program. The genetic algorithm‐multiple linear regression (GA‐MLR) and backward methods—were used to select the suitable descriptors and also for predicting the thermodynamic properties of pyridine derivatives. The correlations between the molecular descriptors in the best models were discussed by the Pearson coefficient correlation and collinearity statistics. The predictive powers of the GA‐MLR models are studied using leave‐one‐out (LOO) cross‐validation and external test set. The best quantitative structure–property relationship (QSPR) models are obtained based on the statistical parameters, such as R2, Q2 LOO, and RMSE values for the data sets. The predictive ability of the GA‐MLR models with two‐three selected molecular descriptors was found to be satisfactory and could be used for modeling and predicting of the mentioned properties of nontested pyridine derivatives. [ABSTRACT FROM AUTHOR]

A quantitative structure–property relationship (QSPR) has been widely developed to derive a correlation between chemical structures of molecules and their known properties. In the present study, the applicability of various molecular descriptors is tested for the QSPR study on 60 sulfonamide drugs (sulfa drugs). A training set of 50 sulfa drugs is used to construct QSPR models for the prediction of thermodynamic properties such as enthalpy of formation (ΔHf/kJ mol−1) and Gibbs free energy (ΔGf/kJ mol−1). These properties are analyzed at the B3LYP/6‐31G* level using Gaussian 98 software. The QSPR models were optimized using multiple linear regression (MLR) analysis. The Genetic algorithm and backward method were used to reduce the number of descriptors derived from the Dragon software. The multicollinearity and autocorrelation properties of the descriptors contributed in the models were tested by calculating the variance inflation factor (VIF), Pearson correlation coefficient (PCC), and the Durbin–Watson (DW) statistics. The predictive powers of the models are discussed using leave‐one‐out cross‐validation (LOOCV) and external validation methods. The statistical coefficients of the models were found to be satisfactory. Thus, QSPR models derived from this study may be helpful for designing some new sulfa drugs and for predicting their properties. [ABSTRACT FROM AUTHOR]

A novel quantitative structure‐property relationship (QSPR) modelling method developed for predicting the ignition delay time of 31 hypergolic imidazolium ionic liquid derivatives which is WFNA (white fuming nitric acid) used as their oxidizer. Chemical structures of these ILs were encoded into several sets of molecular descriptors, representing various physical and chemical properties of these molecules. A total number of 1447 molecular descriptor with different types (e.g., constitutional, functional group counts, and geometrical descriptors) were selected and calculated, then the combination of stepwise and best subset regression approaches was used to determine the molecular descriptors which have significant effects on ignition delay time of hypergolic imidazolium ionic liquid derivatives, finally the response plot of PLS regression approach applied to the selected descriptors, to choose the best number of them. The final derived model displayed a good predictive ability arising from the determination coefficient corresponding to the leave one out cross validation technique (R2=0.9925 and QLOO2 =0.9834). The predictive power of the proposed method is evaluated using several different techniques and this method can be used as a reliable tool for designing novel hypergolic ILs. [ABSTRACT FROM AUTHOR]

Controlled drug delivery devices were predicted in a reverse engineering framework for the controlled release of Paclitaxel, an anti‐cancer drug, widely used in the treatment of solid tumors. Using quantitative structure–property relationship models for mutual diffusion coefficients of the drug in biocompatible and biodegradable polymers and partition coefficients of the drug between polymers and blood, a framework was developed to predict optimal drug delivery devices for desired dosage regimens. The validation of the predicted mutual diffusion and partition coefficients using experimental data was reported in previous studies. Optimal design parameters along with selection of most appropriate polymers suitable for different dosage regimens, selected based on current clinical practice, were predicted for maximum bioavailability of the drug while maintaining the released drug concentration in blood within the therapeutic range. Reservoir and monolithic type of diffusion‐controlled drug delivery devices of different shapes and sizes were predicted with different initial drug loadings and bioavailability for different dosage regimens. The effects of the released Paclitaxel from these devices on the tumor growth were also modeled using a previously reported mathematical pharmacokinetic–pharmacodynamic model. The proposed approach can easily be used to design other diffusion‐controlled drug delivery devices. [ABSTRACT FROM AUTHOR]

The acidity of Lewis-Brønsted superacids can be derived from the theoretical calculations as the Gibbs free energy of the deprotonation reaction (ΔGacid), which describes the tendency of a studied compound to donate a proton. This paper presents the first Quantitative Structure -- Property Relationship (QSPR) model that correlates the DGacid of superacid (HF/MeX3 formula (X=F, Cl, Br)) with their structure. Developed model is well fitted, roubustness, has good predictive abilities, fulfills all OECD recommendation for good model. Obtained results provide the insight into the relation of structural features of superacids, which are responsible for their acid strength -- the structures characterized by strong F-Me dative bond (with relatively large vibrational frequency), small positive partial atomic charge on Me central atom, possibly large polarity exhibit large acid strength. Such assumption can be used in the future as valuable information in the process of the designing new, stronger, more effective superacids. [ABSTRACT FROM AUTHOR]

Cross-linked block copolymers are structurally complex, and utilization of traditional methods of molecular representation in chemoinformatics is only of limited applicability. Therefore, we introduced new techniques of structural representation for block copolymers. We developed additive and combinatorial approaches that treat a copolymer as a mixture system. In this approach, DRAGON descriptors are concentration-weighted for all chemicals in the reaction mixture. As a proof of concept, we have studied glass transition temperatures of block copolymers of hydroxyalkyl- and dihydroxyalkyl carbamate terminated poly(dimethylsiloxane) oligomers with poly(-caprolactone) and developed four quantitative structure-property relationships (QSPR) models. The correlation coefficient (R²) for mentioned QSPR models ranges from 0.851 to 0.911 for the training set. In addition to the newly introduced technique we found that the octanol-water partition coefficient and 3D-MoRSE unweighted descriptors were the most important descriptors for the studied property. The results of the study demonstrated that all chemicals in reaction mixture influenced the glass transition temperatures. [ABSTRACT FROM AUTHOR]

Linear quantitative structure-property relationships (QSPRs) for the prediction of diffusion coefficients (log Dp) were developed for organic contaminants in two common passive sampler materials, polydimethylsiloxane (PDMS) and low-density polyethylene (LDPE). Literature data was compiled for both PDMS and LDPE resulting in final data sets of 196 and 79 compounds, respectively. Data sets contained compounds with log Dp values that ranged over about 5 log units and 3 log units for PDMS and LDPE, respectively. The quality of log Dp prediction using either simple molecular descriptors or quantum-chemical based COSMO-RS sigma moment descriptors was compared for both materials. For PDMS, the sigma moment descriptor QSPR had the best predictivity with a correlation coefficient of R²=0.85 and root mean square error (RMSE) of 0.36 for log Dp. The molecular descriptor QSPR resulted in a correlation coefficient of R²=0.78 and RMSE of 0.45 for log Dp. For LDPE, the molecular descriptor QSPR had the best predictivity, with the final correlation coefficient of R²=0.86 and RMSE of 0.21 for log Dp. The sigma moment descriptor QSPR resulted in a correlation coefficient of R²= 0.66 and RMSE of 0.33 for log Dp. The purely electronic structure-based sigma moments are therefore shown to be a viable option for descriptors compared to the more commonly used molecular descriptors for organic contaminants in PDMS. The significance of the descriptors in each QSPR is discussed. [ABSTRACT FROM AUTHOR]

Abstract: Generative Topographic Mapping (GTM) approach was successfully used to visualize, analyze and model the equilibrium constants (KT) of tautomeric transformations as a function of both structure and experimental conditions. The modeling set contained 695 entries corresponding to 350 unique transformations of 10 tautomeric types, for which KT values were measured in different solvents and at different temperatures. Two types of GTM‐based classification models were trained: first, a “structural” approach focused on separating tautomeric classes, irrespective of reaction conditions, then a “general” approach accounting for both structure and conditions. In both cases, the cross‐validated Balanced Accuracy was close to 1 and the clusters, assembling equilibria of particular classes, were well separated in 2‐dimentional GTM latent space. Data points corresponding to similar transformations measured under different experimental conditions, are well separated on the maps. Additionally, GTM‐driven regression models were found to have their predictive performance dependent on different scenarios of the selection of local fragment descriptors involving special marked atoms (proton donors or acceptors). The application of local descriptors significantly improves the model performance in 5‐fold cross‐validation: RMSE=0.63 and 0.82 logKT units with and without local descriptors, respectively. This trend was as well observed for SVR calculations, performed for the comparison purposes. [ABSTRACT FROM AUTHOR]

Abstract: Cationic surfactants and other low molecular weight compounds are known to inhibit nucleation and agglomeration of methane hydrates. In particular, tetralkylammonium salts are kinetic hydrate inhibitors; ie, they reduce the rate of hydrate formation. This work relates to the in‐silico determination of structural features of molecules modulating methane hydrate formation, as found experimentally, and the prediction of novel structures to be tested as candidate inhibitors. Experimental data for each molecule are the amount of absorbed methane. By inserting these numerical values into a chemoinformatic model, it was possible to find a mutual correlation between structural features and inhibition properties. A maximum amount of information is extracted from the structural features and experimental variables, and a model is generated to explain the relationship therebetween. Chemometric analysis was performed by using the software package Volsurf+ with the aim of finding a primary correlation between surfactant structures and their properties. Experimental parameters (pressure, temperature, and concentration) were further processed through an optimization procedure. A careful study of the chemometric analysis responses and the numerical descriptors of tested surfactants allowed to define the features of a good inhibitor, as far as the amount of absorbed gas is concerned. An external prediction is finally made to project external compounds, whose structures and critical micellar concentration are known, in a statistical model, to predict the inhibition properties of a particular molecule in advance of synthesis and testing. This method allowed to find novel amphiphilic molecules for testing as candidate inhibitors in flow‐assurance. [ABSTRACT FROM AUTHOR]

ABSTRACT: The glass transition temperature (Tg) is one of the most important properties affecting the stability of a polymeric material. A cheminformatics‐based approach has been employed to investigate the glass transition temperatures for a set of polymers. Specifically, a set of 80 polymers was used to build a quantitative structure–property relationship (QSAR). By applying a combination of cheminformatics methods, several predictive models were developed consisting of 1–10 physicochemical variables. The best predictive model, which is based on seven descriptors, successfully predicts the glass transition temperatures for the investigated polymers. Furthermore, the best developed model identified several significant descriptors responsible for glass transition temperatures of the investigated polymers with a correlation coefficient of r2 = 0.77. The computational model derived from this study may serve as a powerful tool to predict glass transition temperatures for various polymers. © 2018 Wiley Periodicals, Inc. J. Polym. Sci., Part B: Polym. Phys. 2018, 56, 877–885 [ABSTRACT FROM AUTHOR]

Abstract: Quantitative structure‐activity/property/toxicity relationship (QSAR/QSPR/QSTR) models are effectively employed to fill data gaps by predicting a given response from known structural features or physicochemical properties of new query compounds. The performance of a model should be assessed based on the quality of predictions checked through diverse validation metrics, which confirm the reliability of the developed QSAR models along with the acceptability of their prediction quality for untested compounds. There is an ongoing effort by QSAR modelers to improve the quality of predictions by lowering the predicted residuals for query compounds. In this endeavor, consensus models integrating all validated individual models were found to be more externally predictive than individual models in many previous studies. The objective of this work has been to explore whether the quality of predictions of external compounds can be enhanced through an “intelligent” selection of multiple models. The consensus predictions used in this study are not simple average of predictions from multiple models. It has been considered in the present study that a particular QSAR model may not be equally effective for prediction of all query compounds in the list. Our approach is different from the previous ones in that none of the previously reported methods considered selection of predictive models in a query compound specific way while at the same time using all or most of the valid models for the total set of query chemicals. We have implemented our approach in a software tool that is freely available via the web http://teqip.jdvu.ac.in/QSAR_Tools/ and http://dtclab.webs.com/software‐tools. [ABSTRACT FROM AUTHOR]

Abstract: Despite the need of a reliable technology for solar energy harvesting, research on new materials for third generation photovoltaics is slowed down by the diffuse use of trial and error rather than rational material design approaches. The proposed study investigates the use of alternative strategies to material discovery inspired by drug design and molecular modeling. In particular, training set and test set (for validation purposes) comprising well‐known small molecule‐bulk heterojunction organic photovoltaics were built. Molecules were characterized by semiempirical calculated and 3D molecular interaction fields–based descriptors. Then partial least squares algorithm was applied to rationalize structure‐photovoltaic activity relationships, and coefficients were investigated to clarify contributions played by the different molecular properties to the final performance. In addition, a photovoltaic desirability function (PhotD) was also proposed as alternative and versatile novel tool for ranking potential candidates. The partial least squares model and PhotD function were both internally and externally validated demonstrating their ability in estimating new candidates performances. The proposed approach demonstrates that, in the context of computational materials science, chemometrics and molecular modeling tools could effectively boost the discovery of novel promising candidates for photovoltaic application. [ABSTRACT FROM AUTHOR]

The relationship between density of energetic azole-based compounds and their molecular structure is investigated through quantitative structure-property relationship (QSPR) approach. The methodology of this work introduces a new model, which related density of azole-based energetic compounds to the optimum elemental composition, the degree of unsaturation (DoU) of the compounds, presence of nitroimino group in the structural formula, as well as several nonadditive structural parameters. The presence of nitroimino functional group and also increasing the value of nO/nN in the formula of these compounds can enhance their density. The correlation is derived on the basis of experimental density values of 100 azole-based energetic compounds with different molecular structure as training set. The determination coefficient of the new correlation is 0.923. Also, it has the root mean square deviation (RMSD) and the average absolute deviation (AAD) of 0.038 and 0.030 g·cm-3, respectively. In addition, the correlation gives good predictions for further 25 azole-based energetic compounds as test set (Q²EXT = 0.901). The predictive ability of the correlation is checked using a cross validation method (Q²LMO = 0.918). The proposed method can also apply for designing novel azolebased energetic compounds. [ABSTRACT FROM AUTHOR]

In drug and material design, the activity and property values of the designed chemical structures can be predicted by quantitative structure−activity and structure−property relationship (QSAR/QSPR) models. When a QSAR/QSPR model is applied to chemical structures, its applicability domain (AD) must be considered. The predicted activity/property values are only reliable for chemical structures inside the AD. Chemical structures outside the AD are usually neglected, as the predicted values are unreliable. The purpose of this study is to develop a methodology for obtaining novel chemical structures with the desired activity or property based on a QSAR/QSPR model by making use of the neglected structures. We propose a structure modification strategy for the AD that considers the activity and property simultaneously. The AD is defined by a one-class support vector machine and the structure modification is guided by a partial derivative of the AD model and matched molecular pairs analysis. Three proof-of-concept case studies generate novel chemical structures inside the AD that exhibit preferable activity/property values according to the QSAR/QSPR model. [ABSTRACT FROM AUTHOR]

In organic chemistry, Comparative Molecular Field Analysis (CoMFA) can be defined as a regression analysis between reaction outcomes and molecular fields, wherein we can extract and visualize important structural information from the coefficients of the constructed regression models. In CoMFA, partial least-squares (PLS) regression, which determines all coefficients in the model, is used for fitting the regression models. However, in organic reactions, steric effects are observed only near the reactive site, indicating that a large number of regression coefficients in the CoMFA of organic reactions should be assigned as 0. The regularized regression method, LASSO/Elastic Net, allows us to fit the regression model while assigning 0 values to unimportant coefficients. Although LASSO/Elastic Net should be suitable for CoMFA, there is no example of its use for organic reaction analysis. Herein, we examine the performance of LASSO/Elastic Net for the quantification of steric effects in CoMFA. We employ digitized molecular structures (the indicator field) as molecular fields that represent steric effects. LASSO/Elastic Net regressions provide highly interpretable models that include less noise than those from PLS regression. © 2017 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

As stricter regulations on CO2 emissions are adopted worldwide, identifying efficient chemical processes to capture and recycle CO2 is of critical importance for industry. The most common process known as amine scrubbing suffers from the lack of available amine solutions capable of capturing CO2 efficiently. Tertiary amines characterized by low heats of reaction are considered good candidates but their absorption properties can significantly differ from one analogue to another despite high structural similarity. Herein, after collecting and curating experimental data from the literature, we have built a modeling set of 41 amine structures with their absorption properties. Then we analyzed their chemical composition using molecular descriptors and non-supervised clustering. Furthermore, we developed a series of quantitative structure-property relationships (QSPR) to assess amines' CO2 absorption properties from their structural characteristics. These models afforded reasonable prediction performances (e. g., Q2LOO=0.63 for CO2 absorption amount) even though they are solely based on 2D chemical descriptors and individual machine learning techniques (random forest and neural network). Overall, we believe the chemical analysis and the series of QSPR models presented in this proof-of-concept study represent new knowledge and innovative tools that could be very useful for screening and prioritizing hypothetical amines to be synthesized and tested experimentally for their CO2 absorption properties. [ABSTRACT FROM AUTHOR]

Thousands of (Quantitative) Structure-Activity Relationships (Q)SAR models have been described in peer-reviewed publications; however, this way of sharing seldom makes models available for the use by the research community outside of the developer's laboratory. Conversely, on-line models allow broad dissemination and application representing the most effective way of sharing the scientific knowledge. Approaches for sharing and providing on-line access to models range from web services created by individual users and laboratories to integrated modeling environments and model repositories. This emerging transition from the descriptive and informative, but 'static', and for the most part, non-executable print format to interactive, transparent and functional delivery of 'living' models is expected to have a transformative effect on modern experimental research in areas of scientific and regulatory use of (Q)SAR models. [ABSTRACT FROM AUTHOR]

In de novo molecular design, quantitative structure-activity relationship (QSAR) and quantitative structure-property relationship (QSPR) models are important to estimate activity and property values, respectively, for virtual molecular structures. To operate QSAR and QSPR models appropriately, applicability domains (ADs) of the models must be defined because estimated values are unreliable for virtual molecular structures that are dissimilar to structures of training compounds. We describe several methods to construct AD models, and then introduce a structure generator to change molecular structures that exist outside ADs to conform to ADs. [ABSTRACT FROM AUTHOR]

In this work, we report QSPR modeling of the free energy Δ G of 1 : 1 hydrogen bond complexes of different H-bond acceptors and donors. The modeling was performed on a large and structurally diverse set of 3373 complexes featuring a single hydrogen bond, for which Δ G was measured at 298 K in CCl4. The models were prepared using Support Vector Machine and Multiple Linear Regression, with ISIDA fragment descriptors. The marked atoms strategy was applied at fragmentation stage, in order to capture the location of H-bond donor and acceptor centers. Different strategies of model validation have been suggested, including the targeted omission of individual H-bond acceptors and donors from the training set, in order to check whether the predictive ability of the model is not limited to the interpolation of H-bond strength between two already encountered partners. Successfully cross-validating individual models were combined into a consensus model, and challenged to predict external test sets of 629 and 12 complexes, in which donor and acceptor formed single and cooperative H-bonds, respectively. In all cases, SVM models outperform MLR. The SVM consensus model performs well both in 3-fold cross-validation ( RMSE=1.50 kJ/mol), and on the external test sets containing complexes with single ( RMSE=3.20 kJ/mol) and cooperative H-bonds ( RMSE=1.63 kJ/mol). [ABSTRACT FROM AUTHOR]

A model developed to predict aqueous solubility at different temperatures has been proposed based on quantitative structure-property relationships (QSPR) methodology. The prediction consists of two steps. The first one predicts the value of k parameter in the linear equation [ABSTRACT FROM AUTHOR]