Quantitative structure–activity relationship prediction of blood-to-brain partitioning behavior using support vector machine

Predictive classification-based read-across for diverse functional vitiligo-linked chemical exposomes (ViCE): A new approach for the assessment of chemical safety for the vitiligo disease in humans

We have explored a new approach using a similarity measure-based read-across derived hypothesis to address the precise risk assessment of vitiligo active chemicals. In this analysis, we initially prepared a data set by combining vitiligo active compounds taken from the previous literature with non-vitiligo chemicals, which are non-skin sensitizers reported in another literature. Afterward, we performed the manual curation process to obtain a curated dataset. Furthermore, the optimum similarity measure was identified from a validation set using a pool of 47 descriptors from the analysis of the most discriminating features. The identified optimum similarity measure (i.e., Euclidean distance-based similarity along with seven close source compounds) has been utilized in the read-across derived similarity-based classification studies on close source congeners concerning target compounds. In this study, we identified the positive and negative contributing features toward the assessment of vitiligo potential as well, including the estimation of target chemicals with better accuracy. The applicability domain status of the reported compounds was also studied, and the outliers were identified. As there are no comparative studies in this regard to the best of our knowledge, we can further affirm that it is the first report on the in-silico identification of potential vitiligo-linked chemical exposomes (ViCE) based on the similarity measure of the read-across.

QSAR modeling on aromatase inhibitory activity of 23 triazole fungicides by tritium-water release assay

The 1,2,4-triazole fungicides are extensively used in agriculture, and their impacts on aquatic organisms by continuous release are increasingly concerned. Aromatase, a rate-limiting enzyme for androgens converting to estrogens, is considered as a potential target for triazole fungicides. To reveal and predict the aromatase inhibition capacity of the existing and future developed triazole fungicides, 23 commonly used 1,2,4-triazole fungicides were used for the evaluation of their inhibitory effects (expressed as the 50% inhibitory concentration (IC50)) on human aromatase by 3H-H2O release assay in the present study. Result showed the IC50 values spanned four orders of magnitude from the strongest of 44 nM (flusilazole) to the lowest of 0.330 mM (bitertanol). The aromatase inhibitory activity of the triazoles was also verified in vivo by zebrafish use two triazoles with relatively weak inhibition. Subsequently, the Quantitative Structure-Activity Relationship (QSAR) modeling on the triazoles as aromatase inhibitors was constructed using stepwise regression analysis with the chemical structural descriptors including physicochemical, electronic and topological parameters. The optimal QSAR model was defined as pIC50 = -22.936-2.668 EHomo + 0.938 logD - 0.715 NHBD. The effectiveness and robustness of the model were evaluated by internal and external validation with residual assessment. The internal validation showed that the R2 and Radj2 were both higher than 0.700. The CCC and CCCExt were in acceptable levels as the cutoff value of 0.850. The cross-validation correlation coefficient Q2 and the external predictive correlation coefficients (Q2-F1, Q2-F2, and Q2-F3) were all greater than 0.600. The results of Y-Scrambling with 2000 iterations indicated the model had no accidental correlation as the average R2 of 0.166 and Q2 of -0.378. The findings offered data support for the potential risks associated with triazole fungicides in aquatic environment and provided theoretical guidance to expedite drug development and risk assessment.

Explainable machine learning models for predicting the acute toxicity of pesticides to sheepshead minnow (Cyprinodon variegatus)

A quantitative structure-activity relationship (QSAR) study was conducted on 313 pesticides to predict their acute toxicity to Sheepshead minnow (Cyprinodon variegatus) by using DRAGON descriptors. Essentials accounting for a reliable model were all considered carefully, giving full consideration to the OECD (Organization for Economic Co-operation and Development) principles for QSAR acceptability in regulation during the model construction and assessment process. Nine variables were selected through the forward stepwise regression method and used as inputs to construct both linear and nonlinear models. The obtained models were validated internally and externally. Generally, machine learning-based methods, namely support vector machine (SVM), random forest (RF), and projection pursuit regression (PPR), perform better than the multiple linear regression (MLR) model. The statistical results (R2 = 0.682-0.933, Q2LOO = 0.604-0.659, Q2F1 = 0.740-0.796, CCC = 0.861-0.882) of the developed models show that they are robust, reliable, reproducible, accurate and predictive. Comparatively, the RF model performs best, giving predictive correlation coefficient Q2 of 0.814, root mean squared error (RMSE) of 0.658 and mean absolute error (MAE) of 0.534 for the test set, respectively. The RF model (as well as SVM and PPR models) was visualized and explained by using the SHapley Additive explanation (SHAP) analysis to enhance its transparency and credibility. In addition, the applicability domain (AD) range of the RF model was characterized by the Williams plot and the tree manifold approximation and projection (TMAP) technology was utilized to illustrate similarity and diversity of the entire data space, to assist in the analysis of the outliers. Activity cliff detection was investigated by using Arithmetic Residuals in K-groups Analysis (ARKA) descriptors. It was found that none of the pesticides was identified as an activity cliff in the training set or a potential prediction cliff in the test set. Therefore, the RF model fulfills each OECD principle in regulation for QSAR models. The research in this work will aid in the in silico QSAR prediction of the acute toxicity to Sheepshead minnow (Cyprinodon variegatus) for untested and new toxic pesticides and can also be extended to other studies.

Elucidation of molecular mechanisms involved in tadpole toxicity employing QSTR and q-RASAR approach

Tadpoles, as early developmental stages of frogs, are vital indicators of toxicity and environmental health in ecosystems exposed to harmful organic compounds from industrial and runoff sources. Evaluating each compound individually is challenging, necessitating the use of in silico methods like Quantitative Structure Toxicity-Relationship (QSTR) and Quantitative Read-Across Structure-Activity Relationship (q-RASAR). Utilizing the comprehensive US EPA's ECOTOX database, which includes acute LC50 toxicity and chronic endpoints, we extracted crucial data such as study types, exposure routes, and chemical categories. Regression-based QSTR and q-RASAR models were developed from this dataset, emphasizing key chemical descriptors. Lipophilicity and unsaturation were significant for predicting acute toxicity, while electrophilicity, nucleophilicity, and molecular branching were crucial for chronic toxicity predictions. Additionally, q-RASAR models integrated with the "intelligent consensus" algorithm were employed to enhance predictive accuracy. The performance of these models was rigorously compared across various approaches. These refined models not only predict the toxicity of untested compounds but also reveal underlying structural influences. Validation through comparison with existing literature affirmed the relevance and robustness of our approach in ecotoxicology.

Safer and greener chemicals for the aquatic ecosystem: Chemometric modeling of the prolonged and chronic aquatic toxicity of chemicals on Oryzias latipes

In the modern era, chemicals and their products have been used everywhere like agriculture, healthcare, food, cosmetics, pharmaceuticals, household products, clothing industry, etc. These chemicals find their way to reach the aquatic ecosystem (directly/indirectly) and cause severe chronic and prolonged toxic effects to aquatic species which is also then translated to human beings. Prolonged and chronic toxicity data of many chemicals that are used daily is not available due to high experimentation testing costs, time investment, and the requirement of a large number of animal sacrifices. Thus, in silico approaches (e.g., QSAR (quantitative structure-activity relationship)) are the best alternative for chronic and prolonged toxicity predictions. The present work offers multi-endpoint (five endpoints: chronic_LOEC, prolonged_14D_LC50, prolonged_14D_NOEC, prolonged_21D_LC50, prolonged_21D_NOEC) QSAR models for addressing the prolonged and chronic aquatic toxicity of chemicals toward fish (O. latipes). The statistical results (R2 =0.738-0.869, QLOO2 =0.712-0.831, Q(F1)2 =0.618-0.731) of the developed models show that they were robust, reliable, reproducible, accurate, and predictive. Some of the features that are responsible for prolonged and chronic toxicity of chemicals towards O. latipes are as follows: the presence of substituted benzene, hydrophobicity, unsaturation, electronegativity, the presence of long-chain fragments, the presence of a greater number of atoms at conjugation, and the presence of halogen atoms. On the other hand, hydrophilicity and graph density descriptors retard the aquatic chronic and prolonged toxicity of chemicals toward O. latipes. The PPDB (pesticide properties database) and experimental and investigational classes of drugs from the DrugBank database were also screened using the developed model. Thus, these multi-endpoint models will be helpful for data-gap filling and provide a broad range of applicability. Therefore, this research will aid in the in silico QSAR (quantitative structure-activity relationship) prediction (non-animal testing) of the prolonged and chronic toxicity of untested and new toxic chemicals/drugs/pesticides, design and development of eco-friendly, novel, and safer chemicals, and help to protect the aquatic ecosystem from exposure to toxic and hazardous chemicals.

Chemometric modeling of pharmaceuticals for partitioning between sludge and aqueous phase during the wastewater treatment process

Computational techniques, such as quantitative structure-property relationships (QSPRs), can play a significant role in exploring the important chemical features essential for the degree of sorption or sludge/water partition coefficient (Kd) towards sewage sludge of wastewater treatment process to evaluate the environmental consequence and risk of pharmaceuticals. The current research work aims to construct a predictive QSPR model for the sorption of 148 diverse active pharmaceutical ingredients (APIs) in sewage sludge during wastewater treatment. For the development of the model, we employed easily computable 2D descriptors as independent variables. The model has been developed following the Organization for Economic Cooperation and Development's (OECD) guidelines. It has undergone internal and external validation using a variety of methodologies, as well as been tested for its applicability domain. A measure of hydrophobicity, i.e., MLOGP2, showed the most promising contribution in modeling the sorption coefficient of APIs. Among other parameters, the number of tertiary aromatic amines, the presence of electronegative atoms like N, O, and Cl, the size of a molecule, the number of aromatic hydroxyl groups, the presence of substituted aromatic nitrogen atoms and alkyl-substituted tertiary carbon atoms were also found to be influential for the regulation of solid water partition coefficient of APIs during the wastewater treatment process. The statistical validity tests performed on the developed partial least squares (PLS) model showed that it is statistically evident, robust, and predictive (R2Train = 0.750, Q2LOO = 0.683, Q2F1 = 0.655, Q2F2 (or R2Test) = 0.651). In addition, the predictivity of the constructed model was further inspected by using the "prediction reliability indicator" tool for 14 external APIs.

Staying Ahead of the Game: How SARS-CoV-2 has Accelerated the Application of Machine Learning in Pandemic Management

In recent years, machine learning (ML) techniques have garnered considerable interest for their potential use in accelerating the rate of drug discovery. With the emergence of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, the utilization of ML has become even more crucial in the search for effective antiviral medications. The pandemic has presented the scientific community with a unique challenge, and the rapid identification of potential treatments has become an urgent priority. Researchers have been able to accelerate the process of identifying drug candidates, repurposing existing drugs, and designing new compounds with desirable properties using machine learning in drug discovery. To train predictive models, ML techniques in drug discovery rely on the analysis of large datasets, including both experimental and clinical data. These models can be used to predict the biological activities, potential side effects, and interactions with specific target proteins of drug candidates. This strategy has proven to be an effective method for identifying potential coronavirus disease 2019 (COVID-19) and other disease treatments. This paper offers a thorough analysis of the various ML techniques implemented to combat COVID-19, including supervised and unsupervised learning, deep learning, and natural language processing. The paper discusses the impact of these techniques on pandemic drug development, including the identification of potential treatments, the understanding of the disease mechanism, and the creation of effective and safe therapeutics. The lessons learned can be applied to future outbreaks and drug discovery initiatives.

Machine learning and artificial intelligence in physiologically based pharmacokinetic modeling

Physiologically based pharmacokinetic (PBPK) models are useful tools in drug development and risk assessment of environmental chemicals. PBPK model development requires the collection of species-specific physiological, and chemical-specific absorption, distribution, metabolism, and excretion (ADME) parameters, which can be a time-consuming and expensive process. This raises a need to create computational models capable of predicting input parameter values for PBPK models, especially for new compounds. In this review, we summarize an emerging paradigm for integrating PBPK modeling with machine learning (ML) or artificial intelligence (AI)-based computational methods. This paradigm includes 3 steps (1) obtain time-concentration PK data and/or ADME parameters from publicly available databases, (2) develop ML/AI-based approaches to predict ADME parameters, and (3) incorporate the ML/AI models into PBPK models to predict PK summary statistics (eg, area under the curve and maximum plasma concentration). We also discuss a neural network architecture "neural ordinary differential equation (Neural-ODE)" that is capable of providing better predictive capabilities than other ML methods when used to directly predict time-series PK profiles. In order to support applications of ML/AI methods for PBPK model development, several challenges should be addressed (1) as more data become available, it is important to expand the training set by including the structural diversity of compounds to improve the prediction accuracy of ML/AI models; (2) due to the black box nature of many ML models, lack of sufficient interpretability is a limitation; (3) Neural-ODE has great potential to be used to generate time-series PK profiles for new compounds with limited ADME information, but its application remains to be explored. Despite existing challenges, ML/AI approaches will continue to facilitate the efficient development of robust PBPK models for a large number of chemicals.

Intelligent consensus predictions of bioconcentration factor of pharmaceuticals using 2D and fragment-based descriptors

Bioconcentration factors (BCFs) are markers of chemical substance accumulation in organisms, and they play a significant role in determining the environmental risk of various chemicals. Experiments to obtain BCFs are expensive and time-consuming; therefore, it is better to estimate BCF early in the chemical development process. The current research aims to evaluate the ecotoxicity potential of 122 pharmaceuticals and identify possible important structural attributes using BCF as the determining feature against a group of fish species. We have calculated the theoretical 2D descriptors from the OCHEM platform and SiRMS descriptor calculating software. The regression-based quantitative structure-property relationship (QSPR) modeling was used to identify the chemical features responsible for acute fish bioconcentration. Multiple models with the "intelligent consensus" algorithm were employed for the regression-based approach improving the predictive ability of the models. To ensure the robustness and interpretability of the developed models, rigorous validation was performed employing various statistical internal and external validation metrics. From the developed models, it can be specified that the presence of large lipophilic and electronegative moieties greatly enhances the bioaccumulative potential of pharmaceuticals, whereas the hydrophilic characteristics have shown a negative impact on BCF. Furthermore, the developed models were employed to screen the DrugBank database (https://go.drugbank.com/) for assessing the BCF properties of the entire database. The evidence acquired from the modeled descriptors might be used for aquatic risk assessment in the future, with the added benefit of providing an early caution of their probable negative impact on aquatic ecosystems for regulatory purposes.

Versatile in silico modelling of microplastics adsorption capacity in aqueous environment based on molecular descriptor and machine learning

To comprehensively evaluate the hazards of microplastics and their coexisting organic pollutants, the sorption capacity of microplastics is a major issue that is quantified through the microplastic-aqueous sorption coefficient (K_d). Almost all quantitative structure-property relationship (QSPR) models that describe K_d apply only to narrow, relatively homogeneous groups of reactants. Herein, non-hybrid QSPR-based models were developed to predict PE-water (K_PE-w), PE-seawater (K_PE-sw), PVC-water (K_PVC-w) and PP-seawater (K_PP-sw) sorption coefficients at different temperatures, with eight machine learning algorithms. Moreover, novel hybrid intelligent models for predicting K_d more accurately were innovatively developed by applying GA, PSO and AdaBoost algorithms to optimize MLP and ELM models. The results indicated that all three optimization algorithms could improve the robustness and predictability of the standalone MLP and ELM models. In all models trained with K_PE-w, K_PE-sw, K_PVC-w and K_PP-sw data sets, GBDT-1 and XGBoost-1 models, MLP-GA-2 and MLP-PSO-2 models, MLR-3 and MLR-4 models performed better in terms of goodness of fit (R_adj²: 0.907-0.999), robustness (Q_BOOT²: 0.900-0.937) and predictability (R_ext²: 0.889-0.970), respectively. Analyzing the descriptors revealed that temperature, lipophilicity, ionization potential and molecular size were correlated closely with the adsorption capacity of microplastics to organic pollutants. The proposed QSPR models may assist in initial environmental exposure assessments without imposing heavy costs in the early experimental phase.

Versatile in silico modeling of XAD-air partition coefficients for POPs based on abraham descriptor and temperature

The concentration of persistent organic pollutants (POPs) makes remarkable difference to environmental fate. In the field of passive sampling, the partition coefficients between polystyrene-divinylbenzene resin (XAD) and air (i.e., K_XAD-A) are indispensable to obtain POPs concentration, and the K_XAD-A is generally thought to be governed by temperature and molecular structure of POPs. However, experimental determination of K_XAD-A is unrealistic for countless and novel chemicals. Herein, the Abraham solute descriptors of poly parameter linear free energy relationship (pp-LFER) and temperature were utilized to develop models, namely pp-LFER-T, for predicting K_XAD-A values. Two linear (MLR and LASSO) and four nonlinear (ANN, SVM, kNN and RF) machine learning algorithms were employed to develop models based on a data set of 307 sample points. For the aforementioned six models, R²_adj and Q²_ext were both beyond 0.90, indicating distinguished goodness-of-fit and robust generalization ability. By comparing the established models, the best model was observed as the RF model with R²_adj = 0.991, Q²_ext = 0.935, RMSE_tra = 0.271 and RMSE_ext = 0.868. The mechanism interpretation revealed that the temperature, size of molecules and dipole-type interactions were the predominant factors affecting K_XAD-A values. Concurrently, the developed models with the broad applicability domain provide available tools to fill the experimental data gap for untested chemicals. In addition, the developed models were helpful to preliminarily evaluate the environmental ecological risk and understand the adsorption behavior of POPs between XAD membrane and air.

Prediction reliability of QSAR models: an overview of various validation tools

The reliability of any quantitative structure-activity relationship (QSAR) model depends on multiple aspects such as the accuracy of the input dataset, selection of significant descriptors, the appropriate splitting process of the dataset, statistical tools used, and most notably on the measures of validation. Validation, the most crucial step in QSAR model development, confirms the reliability of the developed QSAR models and the acceptability of each step in the model development. The present review deals with various validation tools that involve multiple techniques that improve the model quality and robustness. The double cross-validation tool helps in building improved quality models using different combinations of the same training set in an inner cross-validation loop. This exhaustive method is also integrated for small datasets (< 40 compounds) in another tool, namely the small dataset modeler tool. The main aim of QSAR researchers is to improve prediction quality by lowering the prediction errors for the query compounds. 'Intelligent' selection of multiple models and consensus predictions integrated in the intelligent consensus predictor tool were found to be more externally predictive than individual models. Furthermore, another tool called Prediction Reliability Indicator was explained to understand the quality of predictions for a true external set. This tool uses a composite scoring technique to identify query compounds as 'good' or 'moderate' or 'bad' predictions. We have also discussed a quantitative read-across tool which predicts a chemical response based on the similarity with structural analogues. The discussed tools are freely available from https://dtclab.webs.com/software-tools or http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ and https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home (for read-across).

Polyethylene-water partition coefficients for polychlorinated biphenyls: Application of QSPR predictions models with experimental validation

The water environmental recalcitrance and ecotoxicity caused by polychlorinated biphenyls (PCBs) are international issues of common concern. The partition coefficients with PCBs between low-density polyethylene (LDPE) and water (K_PE-w) are significant to assess their environmental transport and/or fate in aquatic environment. Even moderately hydrophobic PCBs, however, possess large K_PE-w values, which makes directly experimental measurement labored. Here, based on the combination of quantitative structure-property relationships (QSPRs) and machine-learning algorithms, 10 in-silico models are developed to provide a quick estimate of K_PE-w. These models exhibit good goodness-of-fit (R²_adj: 0.919-0.975), robustness (Q²_LOO: 0.870-0.954) and external prediction performances (Q²_ext: 0.880-0.971), providing a speedy feasibility to close data gaps for limited or absent experimental information, especially the RF-2 model. Particularly, an additional experimental verification is performed for models by a rapid and accurate three-phase system (aqueous phase, surfactant micelles and LDPE). The results of the experiments for 16 PCBs show the modeling results agree well with experimental values, within or approaching the residuals of ± 0.3 log unit. Mechanism interpretations imply that the number of chlorine atoms and ortho-substituted chlorines are the great effect parameters for K_PE-w. This result also heightens interest in measuring and predicting the K_PE-w values of chemicals containing halogen atoms in water.

Benchmarking prognosis methods for survivability - A case study for patients with contingent primary cancers

BACKGROUND

There is an increasing number of patients with a first primary cancer who are diagnosed with a second primary cancer, but prognosis methods to predict the survivability of a patient with multiple primary cancers have not been fully benchmarked.

METHODS

This study investigated the five-year survivability prognosis performances of six machine learning approaches. These approaches are: artificial neural network, decision tree (DT), logistic regression, support vector machine, naïve Bayes (NB), and Bayesian network (BN). A synthetic minority over-sampling technique (SMOTE) was used to solve the imbalanced problem, and a nationwide cancer patient database containing 7,845 subjects in Taiwan was used as a sample source. Ten primary and secondary cancers and their key variables affecting the survivability of the patients were identified.

RESULTS

All the models using SMOTE improved sensitivity and specificity significantly. NB has the highest performance in terms of accuracy and specificity, whereas BN has the highest performance in terms of sensitivity. Further, the computational time and the power of knowledge representation of NB, BN, and DT outperformed the others.

CONCLUSIONS

Selecting the appropriate prognosis models to predict survivability of patients with two contingent primary cancers can aid precise prediction and can support appropriate treatment advice.

QSPR modelling for investigation of different properties of aminoglycoside-derived polymers using 2D descriptors

The quantitative structure-property relationship (QSPR) method is commonly used to predict different physicochemical characteristics of interest of chemical compounds with an objective to accelerate the process of design and development of novel chemical compounds in the biotechnology and healthcare industries. In the present report, we have employed a QSPR approach to predict the different properties of the aminoglycoside-derived polymers (i.e. polymer DNA binding and aminoglycoside-derived polymers mediated transgene expression). The final QSPR models were obtained using the partial least squares (PLS) regression approach using only specific categories of two-dimensional descriptors and subsequently evaluated considering different internationally accepted validation metrics. The proposed models are robust and non-random, demonstrating excellent predictive ability using test set compounds. We have also developed different kinds of consensus models using several validated individual models to improve the prediction quality for external set compounds. The present findings provide new insight for exploring the design of an aminoglycoside-derived polymer library based on different identified physicochemical properties as well as predict their property before their synthesis.

Prediction of the Blood-Brain Barrier (BBB) Permeability of Chemicals Based on Machine-Learning and Ensemble Methods

The ability of chemicals to enter the blood-brain barrier (BBB) is a key factor for central nervous system (CNS) drug development. Although many models for BBB permeability prediction have been developed, they have insufficient accuracy (ACC) and sensitivity (SEN). To improve performance, ensemble models were built to predict the BBB permeability of compounds. In this study, in silico ensemble-learning models were developed using 3 machine-learning algorithms and 9 molecular fingerprints from 1757 chemicals (integrated from 2 published data sets) to predict BBB permeability. The best prediction performance of the base classifier models was achieved by a prediction model based on an random forest (RF) and a MACCS molecular fingerprint with an ACC of 0.910, an area under the receiver-operating characteristic (ROC) curve (AUC) of 0.957, a SEN of 0.927, and a specificity of 0.867 in 5-fold cross-validation. The prediction performance of the ensemble models is better than that of most of the base classifiers. The final ensemble model has also demonstrated good accuracy for an external validation and can be used for the early screening of CNS drugs.

Artificial intelligence and machine learning-aided drug discovery in central nervous system diseases: State-of-the-arts and future directions

Neurological disorders significantly outnumber diseases in other therapeutic areas. However, developing drugs for central nervous system (CNS) disorders remains the most challenging area in drug discovery, accompanied with the long timelines and high attrition rates. With the rapid growth of biomedical data enabled by advanced experimental technologies, artificial intelligence (AI) and machine learning (ML) have emerged as an indispensable tool to draw meaningful insights and improve decision making in drug discovery. Thanks to the advancements in AI and ML algorithms, now the AI/ML-driven solutions have an unprecedented potential to accelerate the process of CNS drug discovery with better success rate. In this review, we comprehensively summarize AI/ML-powered pharmaceutical discovery efforts and their implementations in the CNS area. After introducing the AI/ML models as well as the conceptualization and data preparation, we outline the applications of AI/ML technologies to several key procedures in drug discovery, including target identification, compound screening, hit/lead generation and optimization, drug response and synergy prediction, de novo drug design, and drug repurposing. We review the current state-of-the-art of AI/ML-guided CNS drug discovery, focusing on blood-brain barrier permeability prediction and implementation into therapeutic discovery for neurological diseases. Finally, we discuss the major challenges and limitations of current approaches and possible future directions that may provide resolutions to these difficulties.

Quantitative structure-property relationships for the calculation of the soil adsorption coefficient using machine learning algorithms with calculated chemical properties from open-source software

The soil adsorption coefficient (K_oc) is an environmental fate parameter that is essential for environmental risk assessment. However, obtaining K_oc requires a significant amount of time and enormous expenditure. Thus, it is necessary to efficiently estimate K_oc in the early stages of a chemical's development. In this study, a quantitative structure-property relationship (QSPR) model was developed using calculated physicochemical properties and molecular descriptors with the OPEn structure-activity/property Relationship App (OPERA) and Mordred software using the largest available K_oc dataset. Specifically, we compared the accuracies of the model using the light gradient boosted machine (LightGBM), a gradient boosting decision tree (GBDT) algorithm, with those of previous models. The experimental results suggested the potential to develop a QSPR model that will produce highly accurate K_oc values using molecular descriptors and physicochemical properties. Unlike previous studies, the use of a combination of LightGBM, OPERA and Mordred enables the prediction of K_oc for many chemicals with high accuracy. In this study, OPERA was used to calculate the physicochemical properties, and Mordred was used to calculate molecular descriptors. The wide range of chemicals covered by OPERA and Mordred enables the analysis of a diverse range of chemical compounds. We also report a method to tune the LightBGM program. The use of fast-processing software, such as LightGBM, enables parameter tuning of a method required to obtain best performance. Our research represents one of the few studies in the field of environmental chemistry to use LightGBM. Using physicochemical properties as well as molecular descriptors, we could develop highly accurate K_oc prediction models when compared to prior studies. In addition, our QSPR models may be useful for preliminary environmental risk assessment without incurring significant costs during the early chemical developmental stage.

Molecular modeling and design of some β-amino alcohol grafted 1,4,5-trisubstituted 1,2,3-triazoles derivatives against chloroquine sensitive, 3D7 strain of Plasmodium falciparum

Resistance nature of Plasmodium falciparum (P. falciparum) to the most effective antimalarial drug, Artemisinin, intimidate the global goal of total eradication of malarial. In an attempt to overcome this challenge, the research was aimed at designing derivatives of β-amino alcohol grafted 1,4,5-trisubstituted 1,2,3-triazoles with improve activity against the P. falciparum through structural modifications of the most active compound (design template), and their activity determined using the developed theoretical predictive model. To achieve this, the geometries were optimized via density functional theory (DFT) using B3LYP/6-31G^∗ basis set to generate molecular descriptors for model development. Analysis of the developed model and the descriptors mean effect lead to the design of derivatives with improved activity. Five (5) theoretical models were developed, where the model {pIC₅₀ = 5.95067(SpMin5_Bhi) - 0.0323461(RDF45m) + 0.0203865 (RDF95e) + 0.0499285 (L1m) - 3.50822} with the highest coefficient of determination (R²) of 0.9367, cross-validated R² (Q²cv) of 0.8242, and the external validated R² (R²_pred) of 0.9462, selected as the best model. The mean effect analysis revealed descriptor SpMin5_Bhi as the most contributive. The descriptor encodes the first ionization potentials of the compounds and are influenced by electron-withdrawing/donating substituents. Hence, structural modifications of the compound with the highest activity (a design template) using electron-withdrawing substituents such as –NO₂, –SO₃H, -Br, –I, –CH₂CH₃, and –CH₃ was done at a different positions, to obtain five (5) hypothetical novel compounds. The statistical results, shows the robustness, excellent prediction power, and validity of the selected model. Descriptor analysis revealed the first ionization potential (SpMin5_Bhi) to play a significant role in the activity of β-amino alcohol grafted 1,4,5-trisubstituted 1,2,3-triazoles derivatives. The five design derivatives of β-amino alcohol grafted 1,4,5-trisubstituted 1,2,3-triazoles with higher activities revealed compound 21C to have an antimalarial activity of pIC₅₀ = 6.7573 higher than it co-designed compounds and even the standard drug. This claim could be verified through molecular docking to determine their interaction with the target protein.

First report on chemometric modeling of hydrolysis half-lives of organic chemicals

Hydrolysis is one of the most important processes of transformation of organic chemicals in water. The rates of reactions, final chemical entities of these processes, and half-lives of organic chemicals are of considerable interest to environmental chemists as well as authorities involved in the controlling the processing and disposal of such organic chemicals. In this study, we have proposed QSPR models for the prediction of hydrolysis half-life of organic chemicals as a function of different pH and temperature conditions using only two-dimensional molecular descriptors with definite physicochemical significance. For each model, suitable subsets of variables were elected using a genetic algorithm method; next, the elected subsets of variables were subjected to the best subset selection with a key objective to determine the best combination of descriptors for model generation. Finally, QSPR models were constructed using the best combination of variables employing the partial least squares (PLS) regression technique. Next, every final model was subjected for strict validation employing the internationally accepted internal and external validation parameters. The proposed models could be applicable for data gap filling to determine hydrolysis half-lives of organic chemicals at different environmental conditions. Generally, presence of aliphatic ether and ether functional groups, high percentage of oxygen content in the molecule and presence of O-Si pairs of atoms at topological distance one, results in a shorter hydrolysis half-life of organic chemicals. On the other hand, higher unsaturation content and high percentage of nitrogen content in molecules lead to higher hydrolysis half-life. It is also found that branched and compact molecules will have a lower half-life while straight chain analogues will have a higher half-life. To the best of our knowledge, the presented models are the first reported QSPR models for hydrolysis half-lives of organic chemicals at different pH values.

Survivability modelling using Bayesian network for patients with first and secondary primary cancers

BACKGROUND AND OBJECTIVE

Multiple primary cancers significantly threat patient survivability. Predicting the survivability of patients with two cancers is challenging because its stochastic pattern relates with numerous variables.

METHODS

In this study, a Bayesian network (BN) model was proposed to describe the occurrence of two primary cancers and predict the five-year survivability of patients using probabilistic evidence. Eleven types of major primary cancers and contingent occurrences of secondary cancers were investigated. A nationwide two-cancer database involving 7,845 patients in Taiwan was investigated. The BN topology is rigorously examined and imbalanced dataset is processed by the synthetic minority oversampling technique. The proposed BN survivability prognosis model was compared with benchmark approaches.

RESULTS

The proposed model significantly outperformed the back-propagation neural network, logistic regression, support vector machine, and naïve Bayes in terms of sensitivity, which is a critical performance index for the non-survival group.

CONCLUSIONS

Using the proposed BN model, one can estimate the posterior probabilities for every query provided appropriate prior evidences. The potential survivability information of patients, treatment effects, and socio-demographics factor effects predicted by the proposed model can help in cancer treatment assessment and cancer development monitoring.

QSAR models for the acute toxicity of 1,2,4-triazole fungicides to zebrafish (Danio rerio) embryos

In recent decades, the 1,2,4-triazole fungicides are widely used for crop diseases control, and their toxicity to wild lives and pollution to ecosystem have attracted more and more attention. However, how to quickly and efficiently evaluate the toxicity of these compounds to environmental organisms is still a challenge. In silico method, such like Quantitative Structure-Activity Relationship (QSAR), provides a good alternative to evaluate the environmental toxicity of a large number of chemicals. At the present study, the acute toxicity of 23 1,2,4-triazole fungicides to zebrafish (Danio rerio) embryos was firstly tested, and the LC₅₀ (median lethal concentration) values were used as the bio-activity endpoint to conduct QSAR modelling for these triazoles. After the comparative study of several QSAR models, the 2D-QSAR model was finally constructed using the stepwise multiple linear regression algorithm combining with two physicochemical parameters (logD and μ), an electronic parameter (Q_N1) and a topological parameter (X^vPC₄). The optimal model could be mathematically described as following: pLC₅₀ = -7.24-0.30X^vPC₄ + 0.76logD - 26.15Q_N1 - 0.08μ. The internal validation by leave-one-out (LOO) cross-validation showed that the R²_adj (adjusted noncross-validation squared correlation coefficient), Q² (cross-validation correlation coefficient) and RMSD (root-mean-square error) was 0.88, 0.84 and 0.17, respectively. The external validation indicated the model had a robust predictability with the q² (predictive squared correlation coefficient) of 0.90 when eliminated tricyclazole. The present study provided a potential tool for predicting the acute toxicity of new 1,2,4-triazole fungicides which contained an independent triazole ring group in their molecules to zebrafish embryos, and also provided a reference for the development of more environmentally-friendly 1,2,4-triazole pesticides in the future.

Comparative chemometric and quantitative structure-retention relationship analysis of anisotropic lipophilicity of 1-arylsuccinimide derivatives determined in high-performance thin-layer chromatography system with aprotic solvents

Numerous structurally different amides and imides including succinimide derivatives exhibit diverse bioactive potential. The development of new compounds requires rationalization in the design in order to provide structural changes that guarantee favorable physico-chemical properties, pharmacological activity and safety. In the present research, a comprehensive study with comparison of the chromatographic lipophilicity and other physico-chemical properties of five groups of 1-arylsuccinimide derivatives was conducted. The chemometric analysis of their physico-chemical properties was carried out by using unsupervised (hierarchical cluster analysis and principal component analysis) and supervised pattern recognition methods (linear discriminant analysis), while the correlations between the in silico molecular features and chromatographic lipophilicity were examined applying linear and non-linear Quantitative Structure-Retention Relationship (QSRR) approaches. The main aim of the conducted research was to determine similarities and dissimilarities among the studied 1-arylsuccinimides, to point out the molecular features which have significant influence on their lipophilicity, as well as to establish high-quality QSRR models which can be used in prediction of chromatographic lipophilicity of structurally similar 1-arylsuccinimides. This study is a continuation of analysis and determination of the physico-chemical properties of 1-arylsuccinimides which could be important guidelines in further in vitro and eventually in vivo studies of their biological potential.

Chemometric Modelling of Heat Release Capacity, Total Heat Release and Char Formation of Polymers to Assess Their Flammability Characteristics

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 R 2 (0.802, 0.842 and 0.826), cross-validated leave-one-out Q² (0.759, 0.810 and 0.752) and predictive R² _pred or Q² _ext (0.810, 0.900 and 0.847) for HRC (n_Training =62, n_Test =28), THR (n_Training =64, n_Test =21) and %char (n_Training =49, n_Test =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.

QSAR models for biocides: The example of the prediction of Daphnia magna acute toxicity

Biocides are multi-component products used to control undesired and harmful organisms able to affect human or animal health or to damage natural and manufactured products. Because of their widespread use, aquatic and terrestrial ecosystems could be contaminated by biocides. The environmental impact of biocides is evaluated through eco-toxicological studies with model organisms of terrestrial and aquatic ecosystems. We focused on the development of in silico models for the evaluation of the acute toxicity (EC₅₀) of a set of biocides collected from different sources on the freshwater crustacean Daphnia magna, one of the most widely used model organisms in aquatic toxicology. Toxicological data specific for biocides are limited, so we developed three models for daphnid toxicity using different strategies (linear regression, random forest, Monte Carlo (CORAL)) to overcome this limitation. All models gave satisfactory results in our datasets: the random forest model showed the best results with a determination coefficient r² = 0.97 and 0.89, respectively, for the training (TS) and the validation sets (VS) while linear regression model and the CORAL model had similar but lower performance (r² = 0.83 and 0.75, respectively, for TS and VS in the linear regression model and r² = 0.74 and 0.75 for the CORAL model).

Chemometric modeling to predict air half-life of persistent organic pollutants (POPs)

We have reported here a quantitative structure-property relationship (QSPR) model for prediction of air half-life of organic chemicals using a dataset of 302 diverse organic chemicals employing only two-dimensional descriptors with definite physicochemical meaning in order to avoid the computational complexity for higher dimensional molecular descriptors. The developed model was rigorously validated using the internationally accepted internal and external validation metrics. The final partial least squares (PLS) regression model obtained at three latent variables comprises six simple and interpretable 2D descriptors. The simple and highly robust model with good quality of predictions explains 66% for the variance of the training set (R²) (64% in terms of LOO variance (Q²)) and 76% for test set variance (R²_pred) (prediction quality). This model might be applicable for data gap filling for determination of POPs in the environment, in case of new or untested chemicals falling within the applicability domain of the model. In general, the model indicates that the air half-life of organic chemicals increases with presence of H-bond acceptor atoms, number of halogen atoms and presence of the R-CH-X fragment and lipophilicity, and decreases with presence of a number of halogens on ring C(sp3) (substitution of halogen atoms on a ring).

Application of QSARs in identification of mutagenicity mechanisms of nitro and amino aromatic compounds against Salmonella typhimurium species

In an attempt to describe the underlying causes of mutagenicity mainly due to organic chemicals, quantitative structure-activity relationship (QSAR) models have been developed using two different Salmonella typhimurium mutagenicity endpoints with or without presence of liver metabolic microsomal enzymes (S9) namely TA98-S9 and TA98 + S9. The models were developed using simple 2D variables having definite physicochemical meaning calculated from Dragon, SiRMS, and PaDEL-descriptor software tools. Stepwise regression followed by partial least squares (PLS) regression was used in model development following the strict OECD guidelines for QSAR model development and validation. The models were validated using coefficient of determination R², cross-validation coefficient Q²_LOO (leave one out) while the test set predictions were analyzed using Q²F₁ (coefficient of determination for the test set). Several other internationally accepted validation metrics like MAE_95%train, average r_m(LOO)² and Δr_m(LOO)² (for the training set) were used to check model robustness while predictive efficiency was evaluated using MAE_95%test, average r_m² and Δr_m² (for the test set). The scope of predictions was defined by applicability domain analysis using the DModX approach, a recommended tool for PLS models. The major contributing features related to mutagenicity include lipophilicity, electronegativity, branching and unsaturation, etc. The present manuscript is the first attempt to undertake modeling of two different endpoints (TA98-S9 and TA98 + S9) in order to explore major contributing molecular features linked directly or indirectly to mutagenicity.

Ecotoxicological QSAR modelling of organic chemicals against Pseudokirchneriella subcapitata using consensus predictions approach

The present study provides robust consensus quantitative structure-activity relationship (QSAR) models developed from 334 organic chemicals covering a wide chemical domain for the prediction of effective concentrations of chemicals for 50% and 10% inhibition of algal growth. Only 2D descriptors with definite physicochemical meaning were employed for QSAR model building, whereas development, validation and interpretation were achieved following the strict Organization for Economic Co-operation and Development (OECD) recommended guidelines. Genetic algorithm along with stepwise approach was used in feature selection while the final QSAR models were derived using partial least squares regression technique. The applicability domain of the developed models was also checked. The obtained consensus models were then used to predict 64 organic chemicals having no definite observed responses while the confidence of predictions was checked by the 'prediction reliability indicator' tool. The developed models should be applicable for data gap filling in case of new or untested organic chemicals provided they fall within the domain of the model and can also be implemented to design safer alternatives to the environment.

QSAR modeling of Daphnia magna and fish toxicities of biocides using 2D descriptors

In the recent years, ecotoxicological hazard potential of biocidal products has been receiving increasing attention in the industries and regulatory agencies. Biocides/pesticides are currently one of the most studied groups of compounds, and their registration cannot be done without the empirical toxicity information. In view of limited experimental data available for these compounds, we have developed Quantitative Structure-Activity Relationship (QSAR) models for the toxicity of biocides to fish and Daphnia magna following principles of QSAR modeling recommended by the OECD (Organization for Economic Cooperation and Development). The models were developed using simple and interpretable 2D descriptors and validated using stringent tests. Both models showed encouraging statistical quality in terms of determination coefficient R² (0.800 and 0.648), cross-validated leave-one-out Q² (0.760 and 0.602) and predictive R²_pred or Q²_ext (0.875 and 0.817) for fish (n_Training = 66, n_Test = 22) and Daphnia magna (n_Training = 100, n_Test = 33) toxicity datasets, respectively. These models should be applicable for data gap filling in case of new or untested biocidal compounds falling within the applicability domain of the models. In general, the models indicate that the toxicity increases with lipophilicity and decreases with polarity, branching and unsaturation. We have also developed interspecies toxicity models for biocides using the daphnia and fish toxicity data and used the models for data gap filling.

Ecotoxicological QSAR modeling of organic compounds against fish: Application of fragment based descriptors in feature analysis

Organic compounds (OCs) constitute an enormously large class of highly persistent and toxic chemicals widely used for various purposes throughout the world. Their increased detection in water bodies, mainly sewage treatment plants via industrial discharge, has rendered them to become a cause for ecological concern. The limited availability of experimental toxicological data has necessitated development of models that can help us identify the most hazardous and potentially toxic compounds thus prioritizing the experiments on the selected chemicals. Computational tools such as quantitative structure-activity relationship (QSAR) can be used to predict the missing data and classify the chemicals based on their acute predicted responses for existing as well as not yet synthesized chemicals. In the current study, novel, externally validated, highly robust local QSAR models for different chemical classes and moderately robust global QSAR models were developed using partial least squares (PLS) regression technique using a large dataset of 1121 OCs for the fish mortality endpoint. For feature selection, genetic algorithm along with stepwise regression was used while model validation was performed using various stringent validation criteria following the strict rules of OECD guidelines of QSAR validation. The variables included in the models were obtained from simplex representation of molecular structures (SiRMS) (Version 4.1.2.270), Dragon (Version 7.0) and PaDEL-descriptor software (Version 2.20). The final developed models were robust, externally predictive and characterized by a large chemical as well as biological domain. The predictive efficiency of the developed models was then compared with the ECOSAR tool in order to justify the applicability of the developed models in ecotoxicological predictions for organic chemicals. Better predictive efficiency of the developed QSAR models compared to the ECOSAR derived predictions signifies their applicability in early risk assessment of known as well as untested chemicals in order to design safer alternatives to the environment.

Chemometric modeling of Daphnia magna toxicity of agrochemicals

Over the past few years, the ecotoxicological hazard potential of agrochemicals has received much attention in the industries and regulatory agencies. In the current work, we have developed quantitative structure-activity relationship (QSAR) models for Daphnia magna toxicities of different classes of agrochemicals (fungicides, herbicides, insecticides and microbiocides) individually as well as for the combined set with the application of Organization for Economic Co-operation and Development (OECD) recommended guidelines. The models for the individual data sets as well as for the combined set were generated employing only simple and interpretable two-dimensional descriptors, and subsequently strictly validated using test set compounds. The validated individual models were used to generate consensus models, with the objective to improve the prediction quality and reduced prediction errors. All the individual models of different classes of agrochemicals as well as the global set of agrochemicals showed encouraging statistical quality and prediction ability. The general observations from the derived models suggest that the toxicity increases with lipophilicity and decreases with polarity. The generated models of different classes of agrochemicals and also for the combined set should be applicable for data gap filling for new or untested agrochemical compounds falling within the applicability domain of the developed models.

Ecotoxicological QSAR modeling of endocrine disruptor chemicals

This study reports highly robust externally predictive quantitative structure-toxicity relationship (QSTR) and interspecies quantitative structure-toxicity-toxicity (i-QSTTR) models developed using toxicity data of endocrine disruptor chemicals (EDCs) towards 14 different species falling in four different trophic levels. Genetic algorithm followed by Partial Least Squares (PLS) regression was used in model development following the strict OECD guidelines. The models were developed using 2D descriptors having definite physicochemical meaning and validated by several internationally accepted validation metrics. The scope of predictions was defined by estimating applicability domain of the models. Presence of halogens, sulfur and phosphorus in the molecules greatly influenced the toxicity of EDCs as suggested by continuous repetition of 2D atom pair descriptors. Lipophilic contributions as calculated by logP terms (mainly ALOGP2 and XlogP) were the second most important feature controlling the EDC hazards. Hydrophilic moiety such as functionalities like esters, aliphatic ethers, branching and higher oxygen content reduced the EDC toxicity. Interspecies models were employed in data gap filling following the hierarchy of different species. The reliability of predictions was calculated by the "prediction reliability indicator" tool.

Consensus QSPR modelling for the prediction of cellular response and fibrinogen adsorption to the surface of polymeric biomaterials

In the current study, we have developed predictive quantitative structure-activity relationship (QSAR) models for cellular response (foetal rate lung fibroblast proliferation) and protein adsorption (fibrinogen adsorption (FA)) on the surface of tyrosine-derived biodegradable polymers designed for tissue engineering purpose using a dataset of 66 and 40 biodegradable polymers, respectively, employing two-dimensional molecular descriptors. Best four individual models have been selected for each of the endpoints. These models are developed using partial least squares regression with a unique combination of six and four descriptors for cellular response and protein adsorption, respectively. The generated models were strictly validated using internal and external metrics to determine the predictive ability and robustness of proposed models. Subsequently, the validated individual models for each response endpoints were used for the generation of 'intelligent' consensus models ( http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ ) to improve the quality of predictions for the external data set. These models may help in prediction of virtual polymer libraries for rational design/optimization for properties relevant to biomedical applications prior to their synthesis.

Consensus QSAR modeling of toxicity of pharmaceuticals to different aquatic organisms: Ranking and prioritization of the DrugBank database compounds

In the present work, quantitative structure-activity relationship (QSAR) models have been developed for ecotoxicity of pharmaceuticals on four different aquatic species namely Pseudokirchneriella subcapitata, Daphnia magna, Oncorhynchus mykiss and Pimephales promelas using genetic algorithm (GA) for feature selection followed by Partial Least Squares regression technique according to the Organization for Economic Co-operation and Development (OECD) guidelines. Double cross-validation methodology was employed for selecting suitable models. Only 2D descriptors were used for capturing chemical information and model building, whereas validation of the models was performed by considering various stringent internal and external validation metrics. Interestingly, models could be developed even without using any LogP terms in contrary to the usual dependence of toxicity on lipophilicity. However, the current manuscript proposes highly robust and more predictive models employing computed logP descriptors. The applicability domain study was performed in order to set a predefined chemical zone of applicability for the obtained QSAR models, and the test compounds falling outside the domain were not taken for further analysis while making a prioritized list. An additional comparison was made with ECOSAR, an online expert system for toxicity prediction of organic pollutants, in order to prove predictability of the obtained models. The obtained robust consensus models were utilized to predict the toxicity of a large dataset of approximately 9300 drug-like molecules in order to prioritize the existing drug-like substances in accordance to their acute predicted aquatic toxicities following a scaling technique. Finally, prioritized lists of 500 most toxic chemicals obtained by respective consensus models and those predicted from ECOSAR tool have been reported.

Chemometric modeling of aquatic toxicity of contaminants of emerging concern (CECs) in Dugesia japonica and its interspecies correlation with daphnia and fish: QSTR and QSTTR approaches

The contaminants of emerging concern (CEC) are universally detected in surface water and soil. They can affect the wild life, and their subsequent translocation through the food chain can affect human health, which is an issue of serious concern. Very few amounts of ecotoxicological data are available on the environmental behavior and ecotoxicity of CEC, thus modeling approaches are essential to bridge the existing gap in experimental data. In this present study, we have developed quantitative structure-toxicity relationship (QSTR) models using a data set of 75 compounds for the prediction of aquatic ecotoxicity of CECs on fresh water planarian (Dugesia japonica) by partial least squares (PLS) regression algorithm using simple molecular descriptors selected by genetic algorithm approach. We also explore the correlations between toxicity against D. japonica and those against daphnia (D. magna) and fish (P. promelas), and these were improved on addition of a few molecular descriptors (B08[C-O] and B09[N-O] in case of daphnia and C-006 and H-052 in case of fish) which allowed us to develop predictive interspecies quantitative structure toxicity-toxicity relationship (QSTTR) models, allowing to extrapolate data from one endpoint to another endpoint. The QSTR (Q²_LOO ranging from 0.630 to 0.720 and R²_pred ranging from 0.723 to 0.798) and QSTTR (Q²_LOO = 0.60 and 0.67, R²_pred = 0.88 and 0.84) models have desirable statistical qualities and acceptable internal and external validation measures, meeting rigorous criteria of different validation metrics and showing acceptability for regulatory purposes as proposed by Organization for Economic Cooperation and Development (OECD). Consensus predictions were also performed based on multiple models generated in this study by using the "Intelligent Consensus Predictor" (ICP) tool to enhance the prediction quality for external set compounds.

Development and Validation of a Robust QSAR Model for Benzothiazole Hydrazone Derivatives as Bcl-XL Inhibitors

Background:

B-cell Lymphoma Extra Large (Bcl-XL) belongs to B-cell Lymphoma two (Bcl-2) family. Due to its over-expression and anti-apoptotic role in many cancers, it has been proven to be a more biologically relevant therapeutic target in anti-cancer therapy. In this study, a Quantitative Structure Activity Relationship (QSAR) modeling was performed to establish the link between structural properties and inhibitory potency of benzothiazole hydrazone derivatives against Bcl-XL.

Methods:

The 53 benzothiazole hydrazone derivatives have been used for model development using genetic algorithm and multiple linear regression methods. The data set is divided into training and test set using Kennard-Stone based algorithm. The best QSAR model has been selected with statistically significant r2 = 0.931, F-test =55.488 RMSE = 0.441 and Q2 0.900.

Results:

The model has been tested successfully for external validation (r2 pred = 0.752), as well as different criteria for acceptable model predictability. Furthermore, analysis of the applicability domain has been carried out to evaluate the prediction reliability of external set molecules. The developed QSAR model has revealed that nThiazoles, nROH, EEig13d, WA, BEHv6, HATS6m, RDF035u and IC4 descriptors are important physico-chemical properties for determining the inhibitory activity of these molecules.

Conclusion:

The developed QSAR model is stable for this chemical series, indicating that test set molecules represent the training dataset. The model is statistically reliable with good predictability. The obtained descriptors reflect important structural features required for activity against Bcl-XL. These properties are designated by topology, shape, size, geometry, substitution information of the molecules (nThiazoles and nROH) and electronic properties. In a nutshell, these characteristics can be successfully utilized for designing and screening of novel inhibitors.

Greener chemicals for the future: QSAR modelling of the PBT index using ETA descriptors

Persistent, bioaccumulative and toxic (PBT) chemicals symbolize a group of substances that are not easily degraded; instead, they accumulate in different organisms and exhibit an acute or chronic toxicity. The limited empirical data on PBT chemicals, the high cost of testing together with the regulatory constraints and the international push for reduced animal testing motivate a greater reliance on predictive computational methods like quantitative structure-activity relationship (QSAR) models in PBT assessment. Papa and Gramatica have recently proposed a PBT index that could be computed directly from structural features. In the current study, we have modelled the experimentally derived PBT index data using an extended topological atom (ETA) along with constitutional descriptors to show the usefulness of the ETA indices in modelling the endpoint. The models developed through a double cross-validation (DCV) method gave the best results in terms of both internal and external validation metrics. The developed models were comparable in predictive quality to those previously reported. The current models were further used for consensus predictions of PBT behaviour for a set of pharmaceuticals and a set of synthetic drug-like compounds. The developed models can be used in PBT hazard screening for identification and prioritization of chemicals from the structural information alone.

Improved Prediction of Blood-Brain Barrier Permeability Through Machine Learning with Combined Use of Molecular Property-Based Descriptors and Fingerprints

Blood-brain barrier (BBB) permeability of a compound determines whether the compound can effectively enter the brain. It is an essential property which must be accounted for in drug discovery with a target in the brain. Several computational methods have been used to predict the BBB permeability. In particular, support vector machine (SVM), which is a kernel-based machine learning method, has been used popularly in this field. For SVM training and prediction, the compounds are characterized by molecular descriptors. Some SVM models were based on the use of molecular property-based descriptors (including 1D, 2D, and 3D descriptors) or fragment-based descriptors (known as the fingerprints of a molecule). The selection of descriptors is critical for the performance of a SVM model. In this study, we aimed to develop a generally applicable new SVM model by combining all of the features of the molecular property-based descriptors and fingerprints to improve the accuracy for the BBB permeability prediction. The results indicate that our SVM model has improved accuracy compared to the currently available models of the BBB permeability prediction.

In silico modelling of azole derivatives with tyrosinase inhibition ability: Application of the models for activity prediction of new compounds

Tyrosinase is a metal containing multifunctional enzymes found in animals, fruits and vegetables and constitutes the primary cause for diseases resulting from overproduction of melanin as well as for browning of fruits. Inhibitors of the enzyme have thus gained increased importance in food and cosmetic industry. In the present work, a group of azole derivatives with tyrosinase inhibitory activity were explored to analyse the prime structural attributes of the potent inhibitors. In silico models have been developed in order to have a close insight regarding features of the molecular fragments that may affect the activity of the molecules conducively. The biological pharmacophore of the inhibitors that accounts for their interaction with the tyrosinase enzyme has been ascertained based on the development of a 3D pharmacophore model. The models thus developed were subsequently utilised for screening a set of compounds that were previously synthesised in-house and were reported to possess antioxidant activity. The final selection of active molecules in the screening process was done based on the docking interactions of the molecules with the tyrosinase enzyme and assessment of their degree of binding to the protein. Thus the developed models have been successfully utilised for identifying active compounds from a series of untested molecules.

Prediction of acute toxicity of emerging contaminants on the water flea Daphnia magna by Ant Colony Optimization-Support Vector Machine QSTR models

According to the European REACH Directive, the acute toxicity towards Daphnia magna should be assessed for any industrial chemical with a market volume of more than 1 t/a. Therefore, it is highly recommended to determine the toxicity at a certain confidence level, either experimentally or by applying reliable prediction models. To this end, a large dataset was compiled, with the experimental acute toxicity values (pLC₅₀) of 1353 compounds in Daphnia magna after 48 h of exposure. A novel quantitative structure-toxicity relationship (QSTR) model was developed, using Ant Colony Optimization (ACO) to select the most relevant set of molecular descriptors, and Support Vector Machine (SVM) to correlate the selected descriptors with the toxicity data. The proposed model showed high performance (Q_LOO² = 0.695, R_fitting² = 0.920 and R_test² = 0.831) with low root mean square errors of 0.498 and 0.707 for the training and test set, respectively. It was found that, in addition to hydrophobicity, polarizability and summation of solute-hydrogen bond basicity affected toxicity positively, while minimum atom-type E-state of -OH influenced toxicity values in Daphnia magna inversely. The applicability domain of the proposed model was carefully studied, considering the effect of chemical structure and prediction error in terms of leverage values and standardized residuals. In addition, a new method was proposed to define the chemical space failure for a compound with unknown toxicity to avoid using these prediction results. The resulting ACO-SVM model was successfully applied on an additional evaluation set and the prediction results were found to be very accurate for those compounds that fall inside the defined applicability domain. In fact, compounds commonly found to be difficult to predict, such as quaternary ammonium compounds or organotin compounds were outside the applicability domain, while five representative homologues of LAS (non-ionic surfactants) were, on average, well predicted within one order of magnitude.

QSPR studies for predicting polarity parameter of organic compounds in methanol using support vector machine and enhanced replacement method

In the present work, enhanced replacement method (ERM) and support vector machine (SVM) were used for quantitative structure-property relationship (QSPR) studies of polarity parameter (p) of various organic compounds in methanol in reversed phase liquid chromatography based on molecular descriptors calculated from the optimized structures. Diverse kinds of molecular descriptors were calculated to encode the molecular structures of compounds, such as geometric, thermodynamic, electrostatic and quantum mechanical descriptors. The variable selection method of ERM was employed to select an optimum subset of descriptors. The five descriptors selected using ERM were used as inputs of SVM to predict the polarity parameter of organic compounds in methanol. The coefficient of determination, r², between experimental and predicted polarity parameters for the prediction set by ERM and SVM were 0.952 and 0.982, respectively. Acceptable results specified that the ERM approach is a very effective method for variable selection and the predictive aptitude of the SVM model is superior to those obtained by ERM. The obtained results demonstrate that SVM can be used as a substitute influential modeling tool for QSPR studies.