In silico prediction of chemical neurotoxicity using machine learning

NeuTox 2.0: A hybrid deep learning architecture for screening potential neurotoxicity of chemicals based on multimodal feature fusion

Chemically induced neurotoxicity is a critical aspect of chemical safety assessment. Traditional and costly experimental methods call for the development of high-throughput virtual screening. However, the small datasets of neurotoxicity have limited the application of advanced deep learning techniques. The current study developed a hybrid deep learning architecture, NeuTox 2.0, through multimodal feature fusion for enhanced prediction accuracy and generalization ability. We incorporated transfer learning based on self-supervised learning, graph neural networks, and molecular fingerprints/descriptors. Four datasets were used to profile neurotoxicity; these were related to blood-brain barrier permeability, neuronal cytotoxicity, microelectrode array-based neural activity, and mammalian neurotoxicity. Comprehensive performance evaluations demonstrated that NeuTox 2.0 has relatively higher predictive capability across all statistical metrics. Specifically, NeuTox 2.0 exhibits remarkable performance in three of the four datasets. In the BBB dataset, although it does not outperform the PaDEL descriptor model, its performance closely approximates that of the top single-modal model. The ablation experiments indicated that NeuTox 2.0 can learn the deeper structural differences of molecules from various feature extractions and capture complex interactions and mapping relationships between various modalities, thereby improving performance for neurotoxicity prediction. Evaluations of anti-noise ability indicated that NeuTox 2.0 has excellent noise resistance relative to traditional machine learning. We applied the NeuTox 2.0 model to predict the neurotoxicity of 315,790 compounds in the REACH database. The results showed that 701 compounds exhibited potential neurotoxicity in the four neurotoxicity-related predictions. In conclusion, NeuTox 2.0 can be used as an efficient tool for early neurotoxicity screening of environmental chemicals.

Data-driven toxicity prediction in drug discovery: Current status and future directions

Early toxicity assessment plays a vital role in the drug discovery process on account of its significant influence on the attrition rate of candidates. Recently, constant upgrading of information technology has greatly promoted the continuous development of toxicity prediction. To give an overview of the current state of data-driven toxicity prediction, we reviewed relevant studies and summarized them in three main respects: the features and difficulties of toxicity prediction, the evolution of modeling approaches, and the available tools for toxicity prediction. For each part, we expound the research status, existing challenges, and feasible solutions. Finally, several new directions and suggestions for toxicity prediction are also put forward.

Research trends of computational toxicology: a bibliometric analysis

BACKGROUND

Computational toxicology utilizes computer models and simulations to predict the toxicity of chemicals. Bibliometric studies evaluate the impact of scientific research in a specific field.

METHODS

A bibliometric analysis of the computational methods used in toxicity assessment was conducted on the Web of Science between 1977 and 2024 February 12.

RESULTS

Findings of this study showed that computational toxicology has evolved considerably over the years, moving towards more advanced computational methods, including machine learning, molecular docking, and deep learning. Artificial intelligence significantly enhances computational toxicology research by improving the accuracy and efficiency of toxicity predictions.

CONCLUSION

Generally, the study highlighted a significant rise in research output in computational toxicology, with a growing interest in advanced methods and a notable focus on refining predictive models to optimize drug properties using tools like pkCSM for more precise predictions.

New binary mixtures of fungicides against Macrophomina phaseolina: Machine learning-driven QSAR, read-across prediction, and molecular dynamics simulation

Quantitative Structure-Activity Relationship (QSAR) analysis greatly enhances the development and research of pesticides. This study employed Multiple Linear Regression (MLR), machine learning (ML), and read-across (RA) approaches to investigate the combined effects of binary mixtures of fungicides on Macrophomina phaseolina. Using the Fixed Ratio Ray Design (FRRD) method, 75 binary mixtures of six frequently used fungicides were generated, with many exhibiting additive interactions as indicated by the Concentration Addition (CA) and Independent Action (IA) models. The QSAR analysis revealed that Support Vector Regression (SVR) and Gaussian Process Regression (GPR) models were the most effective, outperforming the Least Squares Kernel (LSK), MLR, and RA methods. SVR achieved an outstanding R2 of 0.95 and Q2LMO of 0.81, whereas GPR demonstrated values of 0.93 and 0.81 for the same metrics. Internal and external validation confirmed the reliability and generalizability of these models, suggesting they could be applied to a wider array of data. Moreover, Molecular Dynamics (MD) simulations showed that the effects of the fungicides are linked to physiological mechanisms rather than intermolecular interactions within their formulations. This study establishes a robust framework for creating potent fungicide combinations that improve disease management efficacy while promoting environmental sustainability and reducing the chemical load to mitigate negative impacts.

Genetic algorithm multiple linear regression and machine learning-driven QSTR modeling for the acute toxicity of sterol biosynthesis inhibitor fungicides

Sterol Biosynthesis Inhibitors (SBIs) are a major class of fungicides used globally. Their widespread application in agriculture raises concerns about potential harm and toxicity to non-target organisms, including humans. To address these concerns, a quantitative structure-toxicity relationship (QSTR) modeling approach has been developed to assess the acute toxicity of 45 different SBIs. The genetic algorithm (GA) was used to identify key molecular descriptors influencing toxicity. These descriptors were then used to build robust QSTR models using multiple linear regression (MLR), support vector regression (SVR), and artificial neural network (ANN) algorithms. The Cross-validation, Y-randomization test, applicability domain methods, and external validation were carried out to evaluate the accuracy and validity of the generated models. The MLR model exhibited satisfactory predictive performance, with an R2 of 0.72. The SVR and ANN models obtained R2 values of 0.7 and 0.8, respectively. ANN model demonstrated superior performance compared to other models, achieving R2 cv and R2 test values of 0.74 and 0.7, respectively. The models passed both internal and external validation, indicating their robustness. These models offer a valuable tool for risk assessment, enabling the evaluation of potential hazards associated with future applications of SBIs.

Recent advances and current challenges of new approach methodologies in developmental and adult neurotoxicity testing

Adult neurotoxicity (ANT) and developmental neurotoxicity (DNT) assessments aim to understand the adverse effects and underlying mechanisms of toxicants on the human nervous system. In recent years, there has been an increasing focus on the so-called new approach methodologies (NAMs). The Organization for Economic Co-operation and Development (OECD), together with European and American regulatory agencies, promote the use of validated alternative test systems, but to date, guidelines for regulatory DNT and ANT assessment rely primarily on classical animal testing. Alternative methods include both non-animal approaches and test systems on non-vertebrates (e.g., nematodes) or non-mammals (e.g., fish). Therefore, this review summarizes the recent advances of NAMs focusing on ANT and DNT and highlights the potential and current critical issues for the full implementation of these methods in the future. The status of the DNT in vitro battery (DNT IVB) is also reviewed as a first step of NAMs for the assessment of neurotoxicity in the regulatory context. Critical issues such as (i) the need for test batteries and method integration (from in silico and in vitro to in vivo alternatives, e.g., zebrafish, C. elegans) requiring interdisciplinarity to manage complexity, (ii) interlaboratory transferability, and (iii) the urgent need for method validation are discussed.

NeuTox: A weighted ensemble model for screening potential neuronal cytotoxicity of chemicals based on various types of molecular representations

Chemical-induced neurotoxicity has been widely brought into focus in the risk assessment of chemical safety. However, the traditional in vivo animal models to evaluate neurotoxicity are time-consuming and expensive, which cannot completely represent the pathophysiology of neurotoxicity in humans. Cytotoxicity to human neuroblastoma cell line (SH-SY5Y) is commonly used as an alternative to animal testing for the assessment of neurotoxicity, yet it is still not appropriate for high throughput screening of potential neuronal cytotoxicity of chemicals. In this study, we constructed an ensemble prediction model, termed NeuTox, by combining multiple machine learning algorithms with molecular representations based on the weighted score of Particle Swarm Optimization. For the test set, NeuTox shows excellent performance with an accuracy of 0.9064, which are superior to the top-performing individual models. The subsequent experimental verifications reveal that 5,5'-isopropylidenedi-2-biphenylol and 4,4'-cyclo-hexylidenebisphenol exhibited stronger SH-SY5Y-based cytotoxicity compared to bisphenol A, suggesting that NeuTox has good generalization ability in the first-tier assessment of neuronal cytotoxicity of BPA analogs. For ease of use, NeuTox is presented as an online web server that can be freely accessed via http://www.iehneutox-predictor.cn/NeuToxPredict/Predict.

Developmental neurotoxicity (DNT) QSAR combination prediction model establishment and structural characteristics interpretation

With the incidence of neurodevelopmental disorders on the rise, it is imperative to screen and evaluate developmental neurotoxicity (DNT) compounds from a large number of environmental chemicals and understand their mechanisms. In this study, DNT qualitative structure-activity relationship (QSAR) study was carried out for the first time based on DNT data of mammals and structural characterization of DNT compounds was preliminarily illustrated. Five different classification algorithms and two feature selection methods were used to construct prediction models. The best model had good predictive ability on the external test set, but a small application domain (AD). Through combining of three different models, both MCC and AD values were improved. Furthermore, electronical properties, van der Waals volume-related properties and S, Cl or P containing substructure were found to be associated with DNT through modeling descriptors analysis and structure alerts (SAs) identification. This study lays a foundation for further DNT prediction of environmental exposures in human and contributes to the understanding of DNT mechanism.

In Silico Prediction of Human Organ Toxicity via Artificial Intelligence Methods

Unpredicted human organ level toxicity remains one of the major reasons for drug clinical failure. There is a critical need for cost-efficient strategies in the early stages of drug development for human toxicity assessment. At present, artificial intelligence methods are popularly regarded as a promising solution in chemical toxicology. Thus, we provided comprehensive in silico prediction models for eight significant human organ level toxicity end points using machine learning, deep learning, and transfer learning algorithms. In this work, our results showed that the graph-based deep learning approach was generally better than the conventional machine learning models, and good performances were observed for most of the human organ level toxicity end points in this study. In addition, we found that the transfer learning algorithm could improve model performance for skin sensitization end point using source domain of in vivo acute toxicity data and in vitro data of the Tox21 project. It can be concluded that our models can provide useful guidance for the rapid identification of the compounds with human organ level toxicity for drug discovery.

High-Resolution Bacterial Cytological Profiling Reveals Intrapopulation Morphological Variations upon Antibiotic Exposure

Phenotypic heterogeneity is crucial to bacterial survival and could provide insights into the mechanism of action (MOA) of antibiotics, especially those with polypharmacological actions. Although phenotypic changes among individual cells could be detected by existing profiling methods, due to the data complexity, only population average data were commonly used, thereby overlooking the heterogeneity. In this study, we developed a high-resolution bacterial cytological profiling method that can capture morphological variations of bacteria upon antibiotic treatment. With an unprecedented single-cell resolution, this method classifies morphological changes of individual cells into known MOAs with an overall accuracy above 90%. We next showed that combinations of two antibiotics induce altered cell morphologies that are either unique or similar to that of an antibiotic in the combinations. With these combinatorial profiles, this method successfully revealed multiple cytological changes caused by a natural product-derived compound that, by itself, is inactive against Acinetobacter baumannii but synergistically exerts its multiple antibacterial activities in the presence of colistin. The findings have paved the way for future single-cell profiling in bacteria and have highlighted previously underappreciated intrapopulation variations caused by antibiotic perturbation.

Machine Learning Modeling and Insights into the Structural Characteristics of Drug-Induced Neurotoxicity

Neurotoxicity can be resulted from many diverse clinical drugs, which has been a cause of concern to human populations across the world. The detection of drug-induced neurotoxicity (DINeurot) potential with biological experimental methods always required a lot of budget and time. In addition, few studies have addressed the structural characteristics of neurotoxic chemicals. In this study, we focused on the computational modeling for drug-induced neurotoxicity with machine learning methods and the insights into the structural characteristics of neurotoxic chemicals. Based on the clinical drug data with neurotoxicity effects, we developed 35 different classifiers by combining five different machine learning methods and seven fingerprint packages. The best-performing model achieved good results on both 5-fold cross-validation (balanced accuracy of 76.51%, AUC value of 0.83, and MCC value of 0.52) and external validation (balanced accuracy of 83.63%, AUC value of 0.87, and MCC value of 0.67). The model can be freely accessed on the web server DINeuroTpredictor (http://dineurot.sapredictor.cn/). We also analyzed the distribution of several key molecular properties between neurotoxic and non-neurotoxic structures. The results indicated that several physicochemical properties were significantly different between the neurotoxic and non-neurotoxic compounds, including molecular polar surface area (MPSA), AlogP, the number of hydrogen bond acceptors (nHAcc) and donors (nHDon), the number of rotatable bonds (nRotB), and the number of aromatic rings (nAR). In addition, 18 structural alerts responsible for chemical neurotoxicity were identified. The structural alerts have been integrated with our web server SApredictor (http://www.sapredictor.cn). The results of this study could provide useful information for the understanding of the structural characteristics and computational prediction for chemical neurotoxicity.

In silico prediction of parkinsonian motor deficits-related neurotoxicants based on the adverse outcome pathway concept

Exposure to neurotoxicants has been associated with Parkinson's disease (PD). Limited by the clinical variation in the signs and symptoms as well as the slow disease progression, the identification of parkinsonian neurotoxicants relies on animal models. Here, we propose an innovative in silico model for the prediction of parkinsonian neurotoxicants. The model was designed based on a validated adverse outcome pathway (AOP) for parkinsonian motor deficits initiated from the inhibition of mitochondrial complex I. The model consists of a molecular docking model for mitochondrial complex I protein to predict the molecular initiating event and a neuronal cytotoxicity Quantitative Structure-Activity Relationships (QSAR) model to predict the cellular outcome of the AOP. Four known PD-related complex I inhibitors and four non-neurotoxic chemicals were utilized to develop the threshold of the models and to validate the model, respectively. The integrated model showed 100% specificity in ruling out the non-neurotoxic chemicals. The screening of 41 neurotoxicants and complex I inhibitors with the model resulted in 16 chemicals predicted to induce parkinsonian disorder through the molecular initiating event of mitochondrial complex I inhibition. Five of them, namely cyhalothrin, deguelin, deltamethrin, diazepam, and permethrin, are cases with direct evidence linking them to parkinsonian motor deficit-related signs and symptoms. The neurotoxicant prediction model for parkinsonian motor deficits based on the AOP concept may be useful in prioritizing chemicals for further evaluations on PD potential.

SApredictor: An Expert System for Screening Chemicals Against Structural Alerts

The rapid and accurate evaluation of chemical toxicity is of great significance for estimation of chemical safety. In the past decades, a great number of excellent computational models have been developed for chemical toxicity prediction. But most machine learning models tend to be “black box”, which bring about poor interpretability. In the present study, we focused on the identification and collection of structural alerts (SAs) responsible for a series of important toxicity endpoints. Then, we carried out effective storage of these structural alerts and developed a web-server named SApredictor (www.sapredictor.cn) for screening chemicals against structural alerts. People can quickly estimate the toxicity of chemicals with SApredictor, and the specific key substructures which cause the chemical toxicity will be intuitively displayed to provide valuable information for the structural optimization by medicinal chemists.

Brain physiome: A concept bridging in vitro 3D brain models and in silico models for predicting drug toxicity in the brain

In the last few decades, adverse reactions to pharmaceuticals have been evaluated using 2D in vitro models and animal models. However, with increasing computational power, and as the key drivers of cellular behavior have been identified, in silico models have emerged. These models are time-efficient and cost-effective, but the prediction of adverse reactions to unknown drugs using these models requires relevant experimental input. Accordingly, the physiome concept has emerged to bridge experimental datasets with in silico models. The brain physiome describes the systemic interactions of its components, which are organized into a multilevel hierarchy. Because of the limitations in obtaining experimental data corresponding to each physiome component from 2D in vitro models and animal models, 3D in vitro brain models, including brain organoids and brain-on-a-chip, have been developed. In this review, we present the concept of the brain physiome and its hierarchical organization, including cell- and tissue-level organizations. We also summarize recently developed 3D in vitro brain models and link them with the elements of the brain physiome as a guideline for dataset collection. The connection between in vitro 3D brain models and in silico modeling will lead to the establishment of cost-effective and time-efficient in silico models for the prediction of the safety of unknown drugs.

Artificial Intelligence-Based Toxicity Prediction of Environmental Chemicals: Future Directions for Chemical Management Applications

Recently, research on the development of artificial intelligence (AI)-based computational toxicology models that predict toxicity without the use of animal testing has emerged because of the rapid development of computer technology. Various computational toxicology techniques that predict toxicity based on the structure of chemical substances are gaining attention, including the quantitative structure-activity relationship. To understand the recent development of these models, we analyzed the databases, molecular descriptors, fingerprints, and algorithms considered in recent studies. Based on a selection of 96 papers published since 2014, we found that AI models have been developed to predict approximately 30 different toxicity end points using more than 20 toxicity databases. For model development, molecular access system and extended-connectivity fingerprints are the most commonly used molecular descriptors. The most used algorithm among the machine learning techniques is the random forest, while the most used algorithm among the deep learning techniques is a deep neural network. The use of AI technology in the development of toxicity prediction models is a new concept that will aid in achieving a scientific accord and meet regulatory applications. The comprehensive overview provided in this study will provide a useful guide for the further development and application of toxicity prediction models.

Leveraging artificial intelligence to advance the understanding of chemical neurotoxicity

Neurotoxicology is a specialty that aims to understand and explain the impact of chemicals, xenobiotics and physical conditions on nervous system function throughout the life span. Herein, we point to the need for integration of novel translational bioinformatics and chemo-informatics approaches, such as machine learning (ML) and artificial intelligence (AI) to the discipline. Specifically, we advance the notion that AI and ML will be helpful in identifying neurotoxic signatures, provide reliable data in predicting neurotoxicity in the context of genetic variability, and improve the understanding of neurotoxic outcomes associated with exposures to mixtures, to name a few.

Prediction of Micronucleus Assay Outcome Using In Vivo Activity Data and Molecular Structure Features

In vivo micronucleus assay is the widely used genotoxic test to determine the extent of chromosomal aberrations caused by the chemicals in human beings, which plays a significant role in the drug discovery paradigm. To reduce the uncertainties of the in vivo experiments and the expenses, we intended to develop novel machine learning-based tools to predict the toxicity of the compounds with high precision. A total of 372 compounds with known toxicity information were retrieved from the PubChem Bioassay database and literature. The fingerprints and descriptors of the compounds were generated using PaDEL and ChemSAR, respectively, for the analysis. The performance of the models was assessed using the three tires of evaluation strategies such as fivefold, tenfold, and validation by external dataset. Further, structural alerts causing genotoxicity of the compounds were identified using SARpy method. Of note, fingerprint-based random forest model built in our analysis is able to demonstrate the highest accuracy of about 0.97 during tenfold cross-validation. In essence, our study highlights that structural alerts such as chlorocyclohexane and trimethylamine are likely to be the leading cause of toxicity in humans. Indeed, we believe that random forest model generated in this study is appropriate for reduction of test animals and should be considered in the future for the good practice of animal welfare.