Quick and efficient quantitative predictions of androgen receptor binding affinity for screening Endocrine Disruptor Chemicals using 2D-QSAR and Chemical Read-Across

Breaking the Barriers: Machine-Learning-Based c-RASAR Approach for Accurate Blood-Brain Barrier Permeability Prediction

The intricate nature of the blood-brain barrier (BBB) poses a significant challenge in predicting drug permeability, which is crucial for assessing central nervous system (CNS) drug efficacy and safety. This research utilizes an innovative approach, the classification read-across structure-activity relationship (c-RASAR) framework, that leverages machine learning (ML) to enhance the accuracy of BBB permeability predictions. The c-RASAR framework seamlessly integrates principles from both read-across and QSAR methodologies, underscoring the need to consider similarity-related aspects during the development of the c-RASAR model. It is crucial to note that the primary goal of this research is not to introduce yet another model for predicting BBB permeability but rather to showcase the refinement in predicting the BBB permeability of organic compounds through the introduction of a c-RASAR approach. This groundbreaking methodology aims to elevate the accuracy of assessing neuropharmacological implications and streamline the process of drug development. In this study, an ML-based c-RASAR linear discriminant analysis (LDA) model was developed using a dataset of 7807 compounds, encompassing both BBB-permeable and -nonpermeable substances sourced from the B3DB database (freely accessible from https://github.com/theochem/B3DB), for predicting BBB permeability in lead discovery for CNS drugs. The model's predictive capability was then validated using three external sets: one containing 276,518 natural products (NPs) from the LOTUS database (accessible from https://lotus.naturalproducts.net/download) for data gap filling, another comprising 13,002 drug-like/drug compounds from the DrugBank database (available from https://go.drugbank.com/), and a third set of 56 FDA-approved drugs to assess the model's reliability. Further diversifying the predictive arsenal, various other ML-based c-RASAR models were also developed for comparison purposes. The proposed c-RASAR framework emerged as a powerful tool for predicting BBB permeability. This research not only advances the understanding of molecular determinants influencing CNS drug permeability but also provides a versatile computational platform for the rapid assessment of diverse compounds, facilitating informed decision-making in drug development and design.

ARKA: a framework of dimensionality reduction for machine-learning classification modeling, risk assessment, and data gap-filling of sparse environmental toxicity data

Due to the lack of experimental toxicity data for environmental chemicals, there arises a need to fill data gaps by in silico approaches. One of the most commonly used in silico approaches for toxicity assessment of small datasets is the Quantitative Structure-Activity Relationship (QSAR), which generates predictive models for the efficient prediction of query compounds. However, the reliability of the predictions from QSARs derived from small datasets is often questionable from a statistical point of view. This is due to the presence of a larger number of descriptors as compared to the number of training compounds, which reduces the degree of freedom of the developed model. To reduce the overall prediction error for a particular QSAR model, we have proposed here the computation of the novel Arithmetic Residuals in K-groups Analysis (ARKA) descriptors. We have reduced the number of modeling descriptors in a supervised manner by partitioning them into K classes (K = 2 here) depending on the higher mean normalized values of the descriptors to a particular response class, thus preventing the loss of chemical information. A scatter plot of the data points using the values of two ARKA descriptors (ARKA_2 vs. ARKA_1) can potentially identify activity cliffs, less confident data points, and less modelable data points. We have used here five representative environmentally relevant endpoints (skin sensitization, earthworm toxicity, milk/plasma partitioning, algal toxicity, and rodent carcinogenicity of hazardous chemicals) with graded responses to which the ARKA framework was applied for classification modeling. On comparing the performance of the models generated using conventional QSAR descriptors and the ARKA descriptors, the prediction quality of the models derived from ARKA descriptors was found, based on multiple graded-data validation metrics-derived decision criteria, much better than the models derived from QSAR descriptors signifying the potential of ARKA descriptors in ecotoxicological classification modeling of small data sets. Additionally, this holds true for the Read-Across approach as well, since the Read-Across predictions using ARKA descriptors supersede the predictions generated from QSAR descriptors. For the ease of users, a Java-based expert system has been developed that computes the ARKA descriptors from the input of QSAR descriptors.

Quantitative Read-Across Structure-Activity Relationship (q-RASAR): A novel approach to estimate the subchronic oral safety (NOAEL) of diverse organic chemicals in rats

We have developed a quantitative safety prediction model for subchronic repeated doses of diverse organic chemicals on rats using the novel quantitative read-across structure-activity relationship (q-RASAR) approach, which uses similarity-based descriptors for predictive model generation. The experimental -Log (NOAEL) values have been used here as a potential indicator of oral subchronic safety on rats as it determines the maximum dose level for which no observed adverse effects of chemicals are found. A total of 186 data points of diverse organic chemicals have been used for the model generation using structural and physicochemical (0D-2D) descriptors. The read-across-derived similarity, error, and concordance measures (RASAR descriptors) have been extracted from the preliminary 0D-2D descriptors. Then, the combined pool of RASAR and the identified 0D-2D descriptors of the training set were employed to develop the final models by using the partial least squares (PLS) algorithm. The developed PLS model was rigorously validated by various internal and external validation metrics as suggested by the Organization for Economic Co-operation and Development (OECD). The final q-RASAR model is proven to be statistically sound, robust and externally predictive (R2=0.85, Q2LOO=0.82 and Q2F1=0.94) that superseding the internal as well as external predictivity of the corresponding quantitative structure-activity relationship (QSAR) model as well as previously reported subchronic repeated dose toxicity model found in the literature. In a nutshell, the q-RASAR is an effective approach that has the potential to be used as a good alternative way to improve external predictivity, interpretability, and transferability for subchronic oral safety prediction as well as ecotoxicity risk identification.

Chemical-induced liver cancer: an adverse outcome pathway perspective

INTRODUCTION

The evaluation of the potential carcinogenicity is a key consideration in the risk assessment of chemicals. Predictive toxicology is currently switching toward non-animal approaches that rely on the mechanistic understanding of toxicity.

AREAS COVERED

Adverse outcome pathways (AOPs) present toxicological processes, including chemical-induced carcinogenicity, in a visual and comprehensive manner, which serve as the conceptual backbone for the development of non-animal approaches eligible for hazard identification. The current review provides an overview of the available AOPs leading to liver cancer and discusses their use in advanced testing of liver carcinogenic chemicals. Moreover, the challenges related to their use in risk assessment are outlined, including the exploitation of available data, the need for semantic ontologies, and the development of quantitative AOPs.

EXPERT OPINION

To exploit the potential of liver cancer AOPs in the field of risk assessment, 3 immediate prerequisites need to be fulfilled. These include developing human relevant AOPs for chemical-induced liver cancer, increasing the number of AOPs integrating quantitative toxicodynamic and toxicokinetic data, and developing a liver cancer AOP network. As AOPs and other areas in the field continue to evolve, liver cancer AOPs will progress into a reliable and robust tool serving future risk assessment and management.

First report on chemometric modeling of tilapia fish aquatic toxicity to organic chemicals: Toxicity data gap filling

The Toxic Substances Control Act (TSCA) mandates the Environmental Protection Agency (EPA) to document chemicals entering the US. Due to the vast range of toxicity endpoints, experimental toxicological study for all chemicals is impossible to conduct. To address this, in silico methods like QSAR and read-across are strategically used to prioritize testing for chemicals lacking ecotoxicity data. Aquatic toxicity is one of the most critical endpoints directly related to aquatic species, mainly fish, followed by direct to indirect effects on humans through drinking water and fish as food, respectively. Therefore, we have employed the ToxValDB database to curate acute LC50 toxicity data for three Tilapia species covering two different genera, an ideal species for aquatic toxicity testing. Employing the curated dataset, we have developed multiple robust and predictive QSAR and quantitative read-across structure-activity relationship (q-RASAR) models for Tilapia zillii, Oreochromis niloticus, and Oreochromis mossambicus which helped to understand the toxicological mode of action (MoA) of the modeled chemicals and predict the aquatic toxicity of new untested chemicals followed by toxicity data gap filling. The best three QSAR models showed encouraging statistical quality in terms of determination coefficient R2 (0.94, 0.74, and 0.77), cross-validated leave-one-out Q2 (0.90, 0.67 and 0.70), and predictive capability in terms of R2pred (0.95, 0.77, and 0.74) for T. zillii, O. niloticus, and O. mossambicus datasets, respectively. The developed best mathematical models were used for the prediction of aquatic toxicity in terms of pLC50 for 297 untested organic chemicals across three major Tilapia species ranging from 1.841 to 8.561 M in terms of environmental risk assessment.

Read-across-based intelligent learning: development of a global q-RASAR model for the efficient quantitative predictions of skin sensitization potential of diverse organic chemicals

Environmental chemicals and contaminants cause a wide array of harmful implications to terrestrial and aquatic life which ranges from skin sensitization to acute oral toxicity. The current study aims to assess the quantitative skin sensitization potential of a large set of industrial and environmental chemicals acting through different mechanisms using the novel quantitative Read-Across Structure-Activity Relationship (q-RASAR) approach. Based on the identified important set of structural and physicochemical features, Read-Across-based hyperparameters were optimized using the training set compounds followed by the calculation of similarity and error-based RASAR descriptors. Data fusion, further feature selection, and removal of prediction confidence outliers were performed to generate a partial least squares (PLS) q-RASAR model, followed by the application of various Machine Learning (ML) tools to check the quality of predictions. The PLS model was found to be the best among different models. A simple user-friendly Java-based software tool was developed based on the PLS model, which efficiently predicts the toxicity value(s) of query compound(s) along with their status of Applicability Domain (AD) in terms of leverage values. This model has been developed using structurally diverse compounds and is expected to predict efficiently and quantitatively the skin sensitization potential of environmental chemicals to estimate their occupational and health hazards.

Prediction-Inspired Intelligent Training for the Development of Classification Read-across Structure-Activity Relationship (c-RASAR) Models for Organic Skin Sensitizers: Assessment of Classification Error Rate from Novel Similarity Coefficients

The advancements in the field of cheminformatics have led to a reduction in animal testing to estimate the activity, property, and toxicity of query chemicals. Read-across structure-activity relationship (RASAR) is an emerging concept that utilizes various similarity functions derived from chemical information to develop highly predictive models. Unlike quantitative structure-activity relationship (QSAR) models, RASAR descriptors of a query compound are computed from its close congeners instead of the compound itself, thus targeting predictions in the model training phase. The objective of the present study is not to propose new QSAR models for skin sensitization but to demonstrate the enhancement in the quality of predictions of the skin-sensitizing potential of organic compounds by developing classification-based RASAR (c-RASAR) models. A diverse, previously curated data set was collected from the literature for which 2D descriptors were computed. The extracted essential features were then used to develop a classification-based linear discriminant analysis (LDA) QSAR model. Furthermore, from the read-across-based predictions, RASAR descriptors were calculated using the basic settings of the hyperparameters for the Laplacian Kernel-based optimum similarity measure. After feature selection, an LDA c-RASAR model was developed, which superseded the prediction quality of the LDA-QSAR model. Various other combinations of RASAR descriptors were also taken to develop additional c-RASAR models, all showing better prediction quality than the LDA QSAR model while using a lower number of descriptors. Various other machine learning c-RASAR models were also developed for comparison purposes. In this work, we have proposed and analyzed three new similarity metrics: gm_class, sm1, and sm2. The first one is an indicator variable used to generate a simple univariate c-RASAR model with good prediction ability, while the remaining two are similarity indices used to analyze possible activity cliffs in the training and test sets and are believed to play an important role in the modelability analysis of data sets.

Modeling and insights into the structural characteristics of endocrine-disrupting chemicals

Endocrine-disrupting chemicals (EDCs) can cause serious harm to human health and the environment; therefore, it is important to rapidly and correctly identify EDCs. Different computational models have been proposed for the prediction of EDCs over the past few decades, but the reported models are not always easily available, and few studies have investigated the structural characteristics of EDCs. In the present study, we have developed a series of artificial intelligence models targeting EDC receptors: the androgen receptor (AR); estrogen receptor (ER); and pregnane X receptor (PXR). The consensus models achieved good predictive results for validation sets with balanced accuracy values of 87.37%, 90.13%, and 79.21% for AR, ER, and PXR binding assays, respectively. Analysis of the physical-chemical properties suggested that several chemical properties were significantly (p < 0.05) different between EDCs and non-EDCs. We also identified structural alerts that can indicate an EDC, which were integrated into the web server SApredictor. These models and structural characteristics can provide useful tools and information in the discrimination and mechanistic understanding of EDCs in drug discovery and environmental risk assessment.

QSAR and Chemical Read-Across Analysis of 370 Potential MGMT Inactivators to Identify the Structural Features Influencing Inactivation Potency

O6-methylguanine-DNA methyltransferase (MGMT) constitutes an important cellular mechanism for repairing potentially cytotoxic DNA damage induced by guanine O6-alkylating agents and can render cells highly resistant to certain cancer chemotherapeutic drugs. A wide variety of potential MGMT inactivators have been designed and synthesized for the purpose of overcoming MGMT-mediated tumor resistance. We determined the inactivation potency of these compounds against human recombinant MGMT using [3H]-methylated-DNA-based MGMT inactivation assays and calculated the IC50 values. Using the results of 370 compounds, we performed quantitative structure-activity relationship (QSAR) modeling to identify the correlation between the chemical structure and MGMT-inactivating ability. Modeling was based on subdividing the sorted pIC50 values or on chemical structures or was random. A total of nine molecular descriptors were presented in the model equation, in which the mechanistic interpretation indicated that the status of nitrogen atoms, aliphatic primary amino groups, the presence of O-S at topological distance 3, the presence of Al-O-Ar/Ar-O-Ar/R..O..R/R-O-C=X, the ionization potential and hydrogen bond donors are the main factors responsible for inactivation ability. The final model was of high internal robustness, goodness of fit and prediction ability (R2pr = 0.7474, Q2Fn = 0.7375-0.7437, CCCpr = 0.8530). After the best splitting model was decided, we established the full model based on the entire set of compounds using the same descriptor combination. We also used a similarity-based read-across technique to further improve the external predictive ability of the model (R2pr = 0.7528, Q2Fn = 0.7387-0.7449, CCCpr = 0.8560). The prediction quality of 66 true external compounds was checked using the "Prediction Reliability Indicator" tool. In summary, we defined key structural features associated with MGMT inactivation, thus allowing for the design of MGMT inactivators that might improve clinical outcomes in cancer treatment.

Quantitative structure activity relationship (QSAR) modeling study of some novel thiazolidine 4-one derivatives as potent anti-tubercular agents

This study aims to develop a QSAR model for Antitubercular activity. The quantitative structure-activity relationship (QSAR) approach predicted the thiazolidine-4-ones derivative's Antitubercular activity. For the QSAR study, 53 molecules with Antitubercular activity on H37Rv were collected from the literature. Compound structures were drawn by ACD/Labs ChemSketch. The energy minimization of the 2D structure was done using the MM2 force field in Chem3D pro. PaDEL Descriptor software was used to construct the molecular descriptors. QSARINS software was used in this work to develop the 2D QSAR model. A series of thiazolidine 4-one with MIC data were taken from the literature to develop the QSAR model. These compounds were split into a training set (43 compounds) and a test set (10 compounds). The PaDEL software calculated 2300 descriptors for this series of thiazolidine 4-one derivatives. The best predictive Model 4, which has R2 of 0.9092, R2adj of 0.8950 and LOF parameter of 0.0289 identify a preferred fit. The QSAR study resulted in a stable, predictive, and robust model representing the original dataset. In the QSAR equation, the molecular descriptor of MLFER_S, GATSe2, Shal, and EstateVSA 6 positively correlated with Antitubercular activity. While the SpMAD_Dzs 6 is negatively correlated with Antitubercular activity. The high polarizability, Electronegativities, Surface area contributions and number of Halogen atoms in the thiazolidine 4-one derivatives will increase the Antitubercular activity.

Quantitative structure activity relationship studies of androgen receptor binding affinity of endocrine disruptor chemicals with index of ideality of correlation, their molecular docking, molecular dynamics and ADME studies

Endocrine disrupter chemicals (EDCs) are both natural and man-made chemicals that mimic, block or interfere with human hormonal system. In the present manuscript, QSAR modeling was performed for the androgen disruptors that interfere with biosynthesis, metabolism or action of androgens that causes adverse effects on male reproductive system. A set of 96 EDCs that exhibited affinity towards androgen receptors (Log RBA) in rats were employed for carrying out QSAR studies using Hybrid descriptors (combination of HFG and SMILES) through Monte Carlo Optimization. Using index of ideality of correlation (TF2), five splits were formed and predictability of five models resulting from these splits was assessed by various validation parameters. Models resulted from first split was the top most one with R2validation = 0.7878. Structural attributes responsible for change in endpoint were studied by employing correlation weights of structural attributes. In order to further validate the model, new EDCs were designed using these attributes. In silico molecular modelling studies were performed to assess the detailed interactions with the receptor. The binding energies of all the designed compounds were observed to be better than lead and are in the range of -10.46 to -14.80. Molecular dynamics simulation of 100 ns was performed for ED01 and NED05. The results revealed that the protein-ligand complex bearing NED05 was more stable than lead ED01 exhibiting better interactions with the receptor. Further, in an attempt to assess their metabolism, ADME studies were evaluated using SwissADME. The developed model enables to predict the characteristics of designed compounds in an authentic way.Communicated by Ramaswamy H. Sarma.

On Some Novel Similarity-Based Functions Used in the ML-Based q-RASAR Approach for Efficient Quantitative Predictions of Selected Toxicity End Points

The novel quantitative read-across structure-activity relationship (q-RASAR) approach uses read-across-derived similarity functions in the quantitative structure-activity relationship (QSAR) modeling framework in a unique way for supervised model generation. The aim of this study is to explore how this workflow enhances the external (test set) prediction quality of conventional QSAR models by the incorporation of some novel similarity-based functions as additional descriptors using the same level of chemical information. To establish this, five different toxicity data sets, for which QSAR models were reported previously, have been considered in the q-RASAR modeling exercise, which uses chemical similarity-derived measures. The identical sets of chemical features along with the same compositions of training and test sets as reported previously were used in the present analysis for ease of comparison. The RASAR descriptors were calculated based on a chosen similarity measure with the default setting of relevant hyperparameter(s) and were then clubbed with the original structural and physicochemical descriptors, and the number of selected features was further optimized by employing a grid search technique applied on the respective training sets. These features were then used to develop multiple linear regression (MLR) q-RASAR models that show enhanced predictivity as compared to the QSAR models developed previously. Moreover, various other ML algorithms like support vector machine (SVM), linear SVM, random forest, partial least squares, and ridge regression were also employed using the same feature combinations as used in the MLR models to compare the prediction qualities. The q-RASAR models for five different data sets possess at least one of the RASAR descriptors, RA function, g_m, and average similarity, suggesting that these are important determinants of similarities that contribute to the development of predictive q-RASAR models, as also evident from the SHAP analysis of the models.

Efficient predictions of cytotoxicity of TiO2-based multi-component nanoparticles using a machine learning-based q-RASAR approach

The availability of experimental nanotoxicity data is in general limited which warrants both the use of in silico methods for data gap filling and exploring novel methods for effective modeling. Read-Across Structure-Activity Relationship (RASAR) is an emerging cheminformatic approach that combines the usefulness of a QSAR model and similarity-based Read-Across predictions. In this work, we have generated simple, interpretable, and transferable quantitative-RASAR (q-RASAR) models which can efficiently predict the cytotoxicity of TiO₂-based multi-component nanoparticles. A data set of 29 TiO₂-based nanoparticles with specific amounts of noble metal precursors was rationally divided into training and test sets, and the Read-Across-based predictions for the test set were generated. The optimized hyperparameters and the similarity approach, which yield the best predictions, were used to calculate the similarity and error-based RASAR descriptors. A data fusion of the RASAR descriptors with the chemical descriptors was done followed by the best subset feature selection. The final set of selected descriptors was used to develop the q-RASAR models, which were validated using the stringent OECD criteria. Finally, a random forest model was also developed with the selected descriptors, which could efficiently predict the cytotoxicity of TiO₂-based multi-component nanoparticles superseding previously reported models in the prediction quality thus showing the merits of the q-RASAR approach. To further evaluate the usefulness of the approach, we have applied the q-RASAR approach also to a second cytotoxicity data set of 34 heterogeneous TiO₂-based nanoparticles which further confirmed the enhancement of external prediction quality of QSAR models after incorporation of RASAR descriptors.

Discovery of novel VEGFR2-TK inhibitors by phthalimide pharmacophore based virtual screening, molecular docking, MD simulation and DFT

Currently, numerous potent chemotherapeutic agents are available in the market but most of them show poor pharmacokinetics, lethal effects and drug resistance during their enduring use. The increased cancer cases, deaths and need of better treatment stimulates us to give newer lifesaving anticancer drugs. The phthalimide derivatives are structurally diverse and exert potential anticancer activity. In this regard, the 3D QSAR Pharmacophore model was developed and validated using fifty-eight phthalimide derivatives. The validation parameters corroborated the reliability and statistical robustness of CEASER Hypo 1. Three databases-NCI Open, Drug Bank, and Asinex were submitted to ADMET and drug-like filtering; 117893 drug-like compounds were mapped on CEASER Hypo 1; and 362 hits with IC50 <1 µM were discovered. These hits were docked on VEGFR2-TK, and in the form of results fifteen hits exhibited greater affinity than sorafenib. The top lead ASN 03206926 was subjected for MD simulation (100 ns) and RMSD, Rg, RMSF, number of hydrogen bonds, and SASA verified that the complex was stable, rigid and highly compact. Results demonstrated GLU885, PHE918, CYS919, LYS920, HIS1026, CYS1045, ASP1046 are the essential residues for favourable interactions. The binding free energy calculations support the affinity and stability revealed by docking and MD simulation. The DFT calculations, negative binding energy and lower HOMO-LUMO band gap revealed that the process is spontaneous and ASN 03206926 is very reactive. Following extensive analysis we suggest that the ASN 03206926 might be employed as a new VEGFR2-TK inhibitor for the treatment of breast and VEGFR2-TK associated cancers.Communicated by Ramaswamy H. Sarma.

Machine Learning-Assisted Prediction of the Biological Activity of Aromatase Inhibitors and Data Mining to Explore Similar Compounds

Designing molecules for drugs has been a hot topic for many decades. However, it is hard and expensive to find a new molecule. Thus, the cost of the final drug is also increased. Machine learning can provide the fastest way to predict the biological activity of druglike molecules. In the present work, machine learning models are trained for the prediction of the biological activity of aromatase inhibitors. Data was collected from the literature. Molecular descriptors are calculated to be used as independent features for model training. The results showed that the R ² values for linear regression, random forest regression, gradient boosting regression, and bagging regression are 0.58, 0.84, 0.77, and 0.80, respectively. Using these models, it is possible to predict the activity of new molecules in a short period of time and at a reasonable cost. Furthermore, Tanimoto similarity is used for similarity analysis, as well as a chemical database is mined to search for similar molecules. Nonetheless, this study provides a framework for repurposing other effective drug molecules to prevent cancer.