Developing QSAR Models with Defined Applicability Domains on PPARγ Binding Affinity Using Large Data Sets and Machine Learning Algorithms

Recognizing high-priority disinfection byproducts based on experimental and predicted endocrine disrupting data: Virtual screening and in vitro study

So far, about 130 disinfection by-products (DBPs) and several DBPs-groups have had their potential endocrine-disrupting effects tested on some endocrine endpoints. However, it is still not clear which specific DBPs, DBPs-groups/subgroups may be the most toxic substances or groups/subgroups for any given endocrine endpoint. In this study, we attempt to address this issue. First, a list of relevant DBPs was updated, and 1187 DBPs belonging to 4 main-groups (aliphatic, aromatic, alicyclic, heterocyclic) and 84 subgroups were described. Then, the high-priority endocrine endpoints, DBPs-groups/subgroups, and specific DBPs were determined from 18 endpoints, 4 main-groups, 84 subgroups, and 1187 specific DBPs by a virtual-screening method. The results demonstrate that most of DBPs could not disturb the endocrine endpoints in question because the proportion of active compounds associated with the endocrine endpoints ranged from 0 (human thyroid receptor beta) to 32% (human transthyretin (hTTR)). All the endpoints with a proportion of active compounds greater than 10% belonged to the thyroid system, highlighting that the potential disrupting effects of DBPs on the thyroid system should be given more attention. The aromatic and alicyclic DBPs may have higher priority than that of aliphatic and heterocyclic DBPs by considering the activity rate and potential for disrupting effects. There were 2 (halophenols and estrogen DBPs), 12, and 24 subgroups that belonged to high, moderate, and low priority classes, respectively. For individual DBPs, there were 23 (2%), 193 (16%), and 971 (82%) DBPs belonging to the high, moderate, and low priority groups, respectively. Lastly, the hTTR binding affinity of 4 DBPs was determined by an in vitro assay and all the tested DBPs exhibited dose-dependent binding potency with hTTR, which was consistent with the predicted result. Thus, more efforts should be performed to reveal the potential endocrine disruption of those high research-priority main-groups, subgroups, and individual DBPs.

Machine learning-enhanced molecular network reveals global exposure to hundreds of unknown PFAS

Unknown forever chemicals like per- and polyfluoroalkyl substances (PFASs) are difficult to identify. Current platforms designed for metabolites and natural products cannot capture the diverse structural characteristics of PFAS. Here, we report an automatic PFAS identification platform (APP-ID) that screens for PFAS in environmental samples using an enhanced molecular network and identifies unknown PFAS structures using machine learning. Our networking algorithm, which enhances characteristic fragment matches, has lower false-positive rate (0.7%) than current algorithms (2.4 to 46%). Our support vector machine model identified unknown PFAS in test set with 58.3% accuracy, surpassing current software. Further, APP-ID detected 733 PFASs in real fluorochemical wastewater, 39 of which are previously unreported in environmental media. Retrospective screening of 126 PFASs against public data repository from 20 countries show PFAS substitutes are prevalent worldwide.

Hepatotoxicity and lipid metabolism disorders of 8:2 polyfluoroalkyl phosphate diester in zebrafish: In vivo and in silico evidence

8:2 polyfluoroalkyl phosphate diester (8:2 diPAP) has been shown to accumulate in the liver, but whether it induces hepatotoxicity and lipid metabolism disorders remains largely unknown. In this study, zebrafish embryos were exposed to 8:2 diPAP for 7 d. Hepatocellular hypertrophy and karyolysis were noted after exposure to 0.5 ng/L 8:2 diPAP, suggesting suppressed liver development. Compared to the water control, 8:2 diPAP led to significantly higher triglyceride and total cholesterol levels, but markedly lower levels of low-density lipoprotein, implying disturbed lipid homeostasis. The levels of two peroxisome proliferator activated receptor (PPAR) subtypes (pparα and pparγ) involved in hepatotoxicity and lipid metabolism were significantly upregulated by 8:2 diPAP, consistent with their overexpression as determined by immunohistochemistry. In silico results showed that 8:2 diPAP formed hydrogen bonds with PPARα and PPARγ. Among seven machine learning models, Adaptive Boosting performed the best in predicting the binding affinities of PPARα and PPARγ on the test set. The predicted binding affinity of 8:2 diPAP to PPARα (7.12) was higher than that to PPARγ (6.97) by Adaptive Boosting, which matched well with the experimental results. Our results revealed PPAR - mediated adverse effects of 8:2 diPAP on the liver and lipid metabolism of zebrafish larvae.

BERT-based language model for accurate drug adverse event extraction from social media: implementation, evaluation, and contributions to pharmacovigilance practices

Introduction

Social media platforms serve as a valuable resource for users to share health-related information, aiding in the monitoring of adverse events linked to medications and treatments in drug safety surveillance. However, extracting drug-related adverse events accurately and efficiently from social media poses challenges in both natural language processing research and the pharmacovigilance domain.

Method

Recognizing the lack of detailed implementation and evaluation of Bidirectional Encoder Representations from Transformers (BERT)-based models for drug adverse event extraction on social media, we developed a BERT-based language model tailored to identifying drug adverse events in this context. Our model utilized publicly available labeled adverse event data from the ADE-Corpus-V2. Constructing the BERT-based model involved optimizing key hyperparameters, such as the number of training epochs, batch size, and learning rate. Through ten hold-out evaluations on ADE-Corpus-V2 data and external social media datasets, our model consistently demonstrated high accuracy in drug adverse event detection.

Result

The hold-out evaluations resulted in average F1 scores of 0.8575, 0.9049, and 0.9813 for detecting words of adverse events, words in adverse events, and words not in adverse events, respectively. External validation using human-labeled adverse event tweets data from SMM4H further substantiated the effectiveness of our model, yielding F1 scores 0.8127, 0.8068, and 0.9790 for detecting words of adverse events, words in adverse events, and words not in adverse events, respectively.

Discussion

This study not only showcases the effectiveness of BERT-based language models in accurately identifying drug-related adverse events in the dynamic landscape of social media data, but also addresses the need for the implementation of a comprehensive study design and evaluation. By doing so, we contribute to the advancement of pharmacovigilance practices and methodologies in the context of emerging information sources like social media.

Predicting the binding configuration and release potential of heavy metals on iron (oxyhydr)oxides: A machine learning study on EXAFS

Heavy metals raise a global concern and can be easily retained by ubiquitous iron (oxyhydr)oxides in natural and engineered systems. The complex interaction between iron (oxyhydr)oxides and heavy metals results in various mineral-metal binding configurations, such as outer-sphere complexes and edge-sharing inner-sphere complexes, which determine the accumulation and release of heavy metals in the environment. However, traditional experimental approaches are time-consuming and inadequate to elucidate the complex binding relationships and configurations between iron (oxyhydr)oxides and heavy metals. Herein, a workflow that integrates the binding configuration data of 11 heavy metals on 7 iron (oxyhydr)oxides and then trains machine learning models to predict unknown binding configurations was proposed. The well-trained multi-grained cascade forest models exhibited high accuracy (> 90%) and predictive performance (R2 ∼ 0.75). The underlying effects of mineral properties, metal ion species, and environmental conditions on mineral-metal binding configurations were fully interpreted with data mining. Moreover, the metal release rate was further successfully predicted based on mineral-metal binding configurations. This work provides a method to accurately and quickly predict the binding configuration of heavy metals on iron (oxyhydr)oxides, which would provide guidance for estimating the potential release behavior of heavy metals and remediating heavy metal pollution in natural and engineered environments.

CatNet: Sequence-based deep learning with cross-attention mechanism for identifying endocrine-disrupting chemicals

Endocrine-disrupting chemicals (EDCs) pose significant environmental and health risks due to their potential to interfere with nuclear receptors (NRs), key regulators of physiological processes. Despite the evident risks, the majority of existing research narrows its focus on the interaction between compounds and the individual NR target, neglecting a comprehensive assessment across the entire NR family. In response, this study assembled a comprehensive human NR dataset, capturing 49,244 interactions between 35,467 unique compounds and 42 NRs. We introduced a cross-attention network framework, "CatNet", innovatively integrating compound and protein representations through cross-attention mechanisms. The results showed that CatNet model achieved excellent performance with an area under the receiver operating characteristic curve (AUCROC) = 0.916 on the test set, and exhibited reliable generalization on unseen compound-NR pairs. A distinguishing feature of our research is its capacity to expand to novel targets. Beyond its predictive accuracy, CatNet offers a valuable mechanistic perspective on compound-NR interactions through feature visualization. Augmenting the utility of our research, we have also developed a graphical user interface, empowering researchers to predict chemical binding to diverse NRs. Our model enables the prediction of human NR-related EDCs and shows the potential to identify EDCs related to other targets.

Black phosphorus quantum dots induce myocardial inflammatory responses and metabolic disorders in mice

As an ultrasmall derivative of black phosphorus (BP) sheets, BP quantum dots (BP-QDs) have been effectively used in many fields. Currently, information on the cardiotoxicity induced by BP-QDs remains limited. We aimed to evaluate BP-QD-induced cardiac toxicity in mice. Histopathological examination of heart tissue sections was performed. Transcriptome sequencing, real-time quantitative PCR (RT‒qPCR), western blotting, and enzyme-linked immunosorbent assay (ELISA) assays were used to detect the mRNA and/or protein expression of proinflammatory cytokines, nuclear factor kappa B (NF-κB), phosphatidylinositol 3 kinase-protein kinase B (PI3K-AKT), peroxisome proliferator-activated receptor gamma (PPARγ), and glucose/lipid metabolism pathway-related genes. We found that heart weight and heart/body weight index (HBI) were significantly reduced in mice after intragastric administration of 0.1 or 1 mg/kg BP-QDs for 28 days. In addition, obvious inflammatory cell infiltration and increased cardiomyocyte diameter were observed in the BP-QD-treated groups. Altered expression of proinflammatory cytokines and genes related to the NF-κB signaling pathway further confirmed that BP-QD exposure induced inflammatory responses. In addition, BP-QD treatment also affected the PI3K-AKT, PPARγ, thermogenesis, oxidative phosphorylation, and cardiac muscle contraction signaling pathways. The expression of genes related to glucose/lipid metabolism signaling pathways was dramatically affected by BP-QD exposure, and the effect was primarily mediated by the PPAR signaling pathway. Our study provides new insights into the toxicity of BP-QDs to human health.

Machine learning coupled with causal inference to identify COVID-19 related chemicals that pose a high concern to drinking water

Various synthetic substances were utilized in large quantities during the recent coronavirus pandemic, COVID-19. Some of these chemicals could potentially enter drinking water sources. Persistent, mobile, and toxic (PMT) substances have been recognized as a threat to drinking water resources. It has not yet been assessed how many COVID-19 related substances could be considered PMT substances. One reason is the lack of high-quality experimental data for the identification of PMT substances. To solve this problem, we applied a machine learning model to identify the PMT substances among COVID-19 related chemicals. The optimal model achieved an accuracy of 90.6% based on external test data. The model interpretation and causal inference indicated that our approach understood causation between PMT properties and molecular descriptors. Notably, the screening results showed that over 60% of the COVID-19 chemicals considered are candidate PMT substances, which should be prioritized to prevent undue pollution of water resources.

In silico prediction and a systematic toxicology-based in vivo investigation uncovering the mechanism of aquatic toxicity caused by beta-lactam antibiotics

Recently, beta-lactam antibiotics have gained attention as significant contributors to public health and environmental issues due to their potential toxicity. Our study employed machine learning to develop a model for assessing the aquatic toxicity of beta-lactam antibiotics on zebrafish. Notably, aztreonam (AZT), a synthetic monobactam and a subclass of beta-lactam antibiotics, demonstrated developmental effects in zebrafish embryos comparable to cephalosporins, indicating a potential for toxicity. Using a systems toxicology-based approach, we identified apoptosis and metabolic disorders as the primary pathways affected by AZT and its impurity F exposure. During the administration of monobactams, we noted that ctsbb, nos2a, and dgat2, genes associated with apoptosis and the metabolic pathway, exhibited significant differential expression. Molecular docking studies were conducted to ascertain the binding affinity between monobactam compounds and their potential targets-Ctsbb, Nos2a, and Dgat2. Furthermore, our research revealed that monobactams influence pre-mRNA alternative splicing, resulting in disruptions in the expression of genes involved in hair cells, brain, spinal cord, and fin regeneration (e.g., krt4, krt5, krt17, cyt1). Notably, we observed a correlation between the levels of rpl3 and rps7 genes, both important ribosomal proteins, and the detected alternative splicing events. Overall, this study enhances our understanding of the toxicity of beta-lactam antibiotics in zebrafish by demonstrating the developmental effects of monobactams and uncovering the underlying mechanisms at the molecular level. It also identifies potential targets for further investigation into the mechanisms of toxicity and provides valuable insights for early assessment of biological toxicity associated with antibiotic pollutants.

Multimodal Model to Predict Tissue-to-Blood Partition Coefficients of Chemicals in Mammals and Fish

Tissue-to-blood partition coefficients (Ptb) are key parameters for assessing toxicokinetics of xenobiotics in organisms, yet their experimental data were lacking. Experimental methods for measuring Ptb values are inefficient, underscoring the urgent need for prediction models. However, most existing models failed to fully exploit Ptb data from diverse sources, and their applicability domain (AD) was limited. The current study developed a multimodal model capable of processing and integrating textual (categorical features) and numerical information (molecular descriptors/fingerprints) to simultaneously predict Ptb values across various species, tissues, blood matrices, and measurement methods. Artificial neural network algorithms with embedding layers were used for the multimodal modeling. The corresponding unimodal models were developed for comparison. Results showed that the multimodal model outperformed unimodal models. To enhance the reliability of the model, a method considering categorical features, weighted molecular similarity density, and weighted inconsistency in molecular activities of structure-activity landscapes was used to characterize the AD. The model constrained by the AD exhibited better prediction accuracy for the validation set, with the determination coefficient, root mean-square error, and mean absolute error being 0.843, 0.276, and 0.213 log units, respectively. The multimodal model coupled with the AD characterization can serve as an efficient tool for internal exposure assessment of chemicals.

Multi-task machine learning models for simultaneous prediction of tissue-to-blood partition coefficients of chemicals in mammals

Tissue-to-blood partition coefficients (Ptb) are crucial for assessing the distribution of chemicals in organisms. Given the lack of experimental data and laborious nature of experimental methods, there is an urgent need to develop efficient predictive models. With the help of machine learning algorithms, i,e., random forest (RF), and artificial neural network (ANN), this study developed multi-task (MT) models that can simultaneously predict Ptb values for various mammalian tissues, including liver, muscle, brain, lung, and adipose. Single-task (ST) models using partial least squares regression, RF, and ANN algorithms for each endpoint were established for comparison. Overall, the performances of MT models were superior to those of ST models. The MT model using ANN algorithms showed the highest prediction accuracy with determination coefficients ranging from 0.704 to 0.886, root mean square errors between 0.223 and 0.410, and mean absolute errors ranging from 0.178 to 0.285 log units. Results showed that lipophilicity and polarizability of molecules significantly influence their partition behavior in organisms. Applicability domains (ADs) of the models were characterized by weighted molecular similarity density, and weighted inconsistency in molecular activities of structure-activity landscapes. When constrained by ADs, the models displayed enhanced predictive accuracy, making them valuable tools for the risk assessment and management of chemicals.

Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery

Molecular recognition is fundamental in biology, underpinning intricate processes through specific protein-ligand interactions. This understanding is pivotal in drug discovery, yet traditional experimental methods face limitations in exploring the vast chemical space. Computational approaches, notably quantitative structure-activity/property relationship analysis, have gained prominence. Molecular fingerprints encode molecular structures and serve as property profiles, which are essential in drug discovery. While two-dimensional (2D) fingerprints are commonly used, three-dimensional (3D) structural interaction fingerprints offer enhanced structural features specific to target proteins. Machine learning models trained on interaction fingerprints enable precise binding prediction. Recent focus has shifted to structure-based predictive modeling, with machine-learning scoring functions excelling due to feature engineering guided by key interactions. Notably, 3D interaction fingerprints are gaining ground due to their robustness. Various structural interaction fingerprints have been developed and used in drug discovery, each with unique capabilities. This review recapitulates the developed structural interaction fingerprints and provides two case studies to illustrate the power of interaction fingerprint-driven machine learning. The first elucidates structure-activity relationships in β2 adrenoceptor ligands, demonstrating the ability to differentiate agonists and antagonists. The second employs a retrosynthesis-based pre-trained molecular representation to predict protein-ligand dissociation rates, offering insights into binding kinetics. Despite remarkable progress, challenges persist in interpreting complex machine learning models built on 3D fingerprints, emphasizing the need for strategies to make predictions interpretable. Binding site plasticity and induced fit effects pose additional complexities. Interaction fingerprints are promising but require continued research to harness their full potential.

Multi-and many-objective optimization: present and future in de novo drug design

de novo Drug Design (dnDD) aims to create new molecules that satisfy multiple conflicting objectives. Since several desired properties can be considered in the optimization process, dnDD is naturally categorized as a many-objective optimization problem (ManyOOP), where more than three objectives must be simultaneously optimized. However, a large number of objectives typically pose several challenges that affect the choice and the design of optimization methodologies. Herein, we cover the application of multi- and many-objective optimization methods, particularly those based on Evolutionary Computation and Machine Learning techniques, to enlighten their potential application in dnDD. Additionally, we comprehensively analyze how molecular properties used in the optimization process are applied as either objectives or constraints to the problem. Finally, we discuss future research in many-objective optimization for dnDD, highlighting two important possible impacts: i) its integration with the development of multi-target approaches to accelerate the discovery of innovative and more efficacious drug therapies and ii) its role as a catalyst for new developments in more fundamental and general methodological frameworks in the field.

Count-Based Morgan Fingerprint: A More Efficient and Interpretable Molecular Representation in Developing Machine Learning-Based Predictive Regression Models for Water Contaminants' Activities and Properties

In this study, we introduce the count-based Morgan fingerprint (C-MF) to represent chemical structures of contaminants and develop machine learning (ML)-based predictive models for their activities and properties. Compared with the binary Morgan fingerprint (B-MF), C-MF not only qualifies the presence or absence of an atom group but also quantifies its counts in a molecule. We employ six different ML algorithms (ridge regression, SVM, KNN, RF, XGBoost, and CatBoost) to develop models on 10 contaminant-related data sets based on C-MF and B-MF to compare them in terms of the model's predictive performance, interpretation, and applicability domain (AD). Our results show that C-MF outperforms B-MF in nine of 10 data sets in terms of model predictive performance. The advantage of C-MF over B-MF is dependent on the ML algorithm, and the performance enhancements are proportional to the difference in the chemical diversity of data sets calculated by B-MF and C-MF. Model interpretation results show that the C-MF-based model can elucidate the effect of atom group counts on the target and have a wider range of SHAP values. AD analysis shows that C-MF-based models have an AD similar to that of B-MF-based ones. Finally, we developed a "ContaminaNET" platform to deploy these C-MF-based models for free use.

A Graph Neural Network Model with a Transparent Decision-Making Process Defines the Applicability Domain for Environmental Estrogen Screening

The application of deep learning (DL) models for screening environmental estrogens (EEs) for the sound management of chemicals has garnered significant attention. However, the currently available DL model for screening EEs lacks both a transparent decision-making process and effective applicability domain (AD) characterization, making the reliability of its prediction results uncertain and limiting its practical applications. To address this issue, a graph neural network (GNN) model was developed to screen EEs, achieving accuracy rates of 88.9% and 92.5% on the internal and external test sets, respectively. The decision-making process of the GNN model was explored through the network-like similarity graphs (NSGs) based on the model features (FT). We discovered that the accuracy of the predictions is dependent on the feature distribution of compounds in NSGs. An AD characterization method called ADFT was proposed, which excludes predictions falling outside of the model's prediction range, leading to a 15% improvement in the F1 score of the GNN model. The GNN model with the AD method may serve as an efficient tool for screening EEs, identifying 800 potential EEs in the Inventory of Existing Chemical Substances of China. Additionally, this study offers new insights into comprehending the decision-making process of DL models.

Advancing food safety risk assessment in China: development of new approach methodologies (NAMs)

Novel techniques and methodologies are being developed to advance food safety risk assessment into the next-generation. Considering the shortcomings of traditional animal testing, new approach methodologies (NAMs) will be the main tools for the next-generation risk assessment (NGRA), using non-animal methodologies such as in vitro and in silico approaches. The United States Environmental Protection Agency and the European Food Safety Authority have established work plans to encourage the development and application of NAMs in NGRA. Currently, NAMs are more commonly used in research than in regulatory risk assessment. China is also developing NAMs for NGRA but without a comprehensive review of the current work. This review summarizes major NAM-related research articles from China and highlights the China National Center for Food Safety Risk Assessment (CFSA) as the primary institution leading the implementation of NAMs in NGRA in China. The projects of CFSA on NAMs such as the Food Toxicology Program and the strategies for implementing NAMs in NGRA are outlined. Key issues and recommendations, such as discipline development and team building, are also presented to promote NAMs development in China and worldwide.

Applicability Domains Based on Molecular Graph Contrastive Learning Enable Graph Attention Network Models to Accurately Predict 15 Environmental End Points

In silico models for predicting physicochemical properties and environmental fate parameters are necessary for the sound management of chemicals. This study employed graph attention network (GAT) algorithms to construct such models on 15 end points. The results showed that the GAT models outperformed the previous state-of-the-art models, and their performance was not influenced by the presence or absence of compounds with certain structures. Molecular similarity density (ρs) was found to be a key metrics characterizing data set modelability, in addition to the proportion of compounds at activity cliffs. By introducing molecular graph (MG) contrastive learning, MG-based ρs and molecular inconsistency in activities (IA) were calculated and employed for characterizing the structure-activity landscape (SAL)-based applicability domain ADSAL{ρs, IA}. The GAT models coupled with ADSAL{ρs, IA} significantly improved the prediction coefficient of determination (R2) on all the end points by an average of 14.4% and enabled all the end points to have R2 > 0.9, which could hardly be achieved previously. The models were employed to screen persistent, mobile, and/or bioaccumulative chemicals from inventories consisting of about 106 chemicals. Given the current state-of-the-art model performance and coverage of the various environmental end points, the constructed models with ADSAL{ρs, IA} may serve as benchmarks for future efforts to improve modeling efficacy.

Predicting adsorption of organic compounds onto graphene and black phosphorus by molecular dynamics and machine learning

With an increase in production and application of various engineering nanomaterials (ENMs), they will inevitably be released into the environment. Adsorption of various organic chemicals onto ENMs will impact on their environmental behavior and toxicology. It is unrealistic to experimentally determine adsorption equilibrium constants (K) for the vast number of organics and ENMs due to high cost in expenditure and time. Herein, appropriate molecular dynamics (MD) methods were evaluated and selected by comparing experimental K values of seven organics adsorbed onto graphene with the MD-calculated ones. Machine learning (ML) models on K of organics adsorption onto graphene and black phosphorus nanomaterials were constructed based on a benchmark data set from the MD simulations. Lasso models based on Mordred descriptors outperformed ML models built by support vector machine, random forest, k-nearest neighbor, and gradient boosting decision tree, in terms of cross-validation coefficients (Q2 > 0.90). The Lasso models also outperformed conventional poly-parameter linear free energy relationship models for predicting logK. Compared with previous models, the Lasso models considered more compounds with different functional groups and thus have broader applicability domains. This study provides a promising way to fill the data gap in logK for chemicals adsorbed onto the ENMs.

Are New Phthalate Ester Substitutes Safer than Traditional DBP and DiBP? Comparative Endocrine-Disrupting Analyses on Zebrafish Using In Vivo, Transcriptome, and In Silico Approaches

Although previous studies have confirmed the association between phthalate esters (PAEs) exposure and endocrine disorders in humans, few studies to date have systematically assessed the threats of new PAE alternatives to endocrine disruptions. Herein, zebrafish embryos were continuously exposed to two PAEs [di-n-butyl phthalate (DBP) and diisobutyl phthalate (DiBP)], two structurally related alternatives [diiononyl phthalate (DINP) and diisononyl hexahydrophthalate (DINCH)], and two non-PAE substitutes [dipropylene glycol dibenzoate (DGD) and glyceryl triacetate (GTA)], and the endocrine-disrupting effects were investigated during the early stages (8-48 hpf). For five endogenous hormones, including progesterone, testosterone, 17β-estradiol, triiodothyronine (T3), and cortisol, the tested chemicals disturbed the contents of at least one hormone at environmentally relevant concentrations (≤3.9 μM), except DINCH and GTA. Then, the concentration-dependent reduced zebrafish transcriptome analysis was performed. Thyroid hormone (TH)- and androgen/estrogen-regulated adverse outcome pathways (AOPs) were the two types of biological pathways most sensitive to PAE exposure. Notably, six compounds disrupted four TH-mediated AOPs, from the inhibition of deiodinases (molecular initiating event, MIE), a decrease in T3 levels (key event, KE), to mortality (adverse outcome, AO) with the quantitatively linear relationships between MIE-KE (|r| = 0.96, p = 0.002), KE-AO (|r| = 0.88, p = 0.02), and MIE-AO (|r| = 0.89, p = 0.02). Multiple structural analyses showed that benzoic acid is the critical toxicogenic fragment. Our data will facilitate the screening and development of green alternatives.

Identification and Prioritization of Environmental Organic Pollutants: From an Analytical and Toxicological Perspective

Exposure to environmental organic pollutants has triggered significant ecological impacts and adverse health outcomes, which have been received substantial and increasing attention. The contribution of unidentified chemical components is considered as the most significant knowledge gap in understanding the combined effects of pollutant mixtures. To address this issue, remarkable analytical breakthroughs have recently been made. In this review, the basic principles on recognition of environmental organic pollutants are overviewed. Complementary analytical methodologies (i.e., quantitative structure-activity relationship prediction, mass spectrometric nontarget screening, and effect-directed analysis) and experimental platforms are briefly described. The stages of technique development and/or essential parts of the analytical workflow for each of the methodologies are then reviewed. Finally, plausible technique paths and applications of the future nontarget screening methods, interdisciplinary techniques for achieving toxicant identification, and burgeoning strategies on risk assessment of chemical cocktails are discussed.

Machine learning-assisted data filtering and QSAR models for prediction of chemical acute toxicity on rat and mouse

Machine learning (ML) methods provide a new opportunity to build quantitative structure-activity relationship (QSAR) models for predicting chemicals' toxicity based on large toxicity data sets, but they are limited in insufficient model robustness due to poor data set quality for chemicals with certain structures. To address this issue and improve model robustness, we built a large data set on rat oral acute toxicity for thousands of chemicals, then used ML to filter chemicals favorable for regression models (CFRM). In comparison to chemicals not favorable for regression models (CNRM), CFRM accounted for 67% of chemicals in the original data set, and had a higher structural similarity and a smaller toxicity distribution in 2-4 log₁₀ (mg/kg). The performance of established regression models for CFRM was greatly improved, with root-mean-square deviations (RMSE) in the range of 0.45-0.48 log₁₀ (mg/kg). Classification models were built for CNRM using all chemicals in the original data set, and the area under receiver operating characteristic (AUROC) reached 0.75-0.76. The proposed strategy was successfully applied to a mouse oral acute data set, yielding RMSE and AUROC in the range of 0.36-0.38 log₁₀ (mg/kg) and 0.79, respectively.

Machine Learning Model for Screening Thyroid Stimulating Hormone Receptor Agonists Based on Updated Datasets and Improved Applicability Domain Metrics

Machine learning (ML) models for screening endocrine-disrupting chemicals (EDCs), such as thyroid stimulating hormone receptor (TSHR) agonists, are essential for sound management of chemicals. Previous models for screening TSHR agonists were built on imbalanced datasets and lacked applicability domain (AD) characterization essential for regulatory application. Herein, an updated TSHR agonist dataset was built, for which the ratio of active to inactive compounds greatly increased to 1:2.6, and chemical spaces of structure-activity landscapes (SALs) were enhanced. Resulting models based on 7 molecular representations and 4 ML algorithms were proven to outperform previous ones. Weighted similarity density (ρ_s) and weighted inconsistency of activities (I_A) were proposed to characterize the SALs, and a state-of-the-art AD characterization methodology AD_SAL{ρ_s, I_A} was established. An optimal classifier developed with PubChem fingerprints and the random forest algorithm, coupled with AD_SAL{ρ_s ≥ 0.15, I_A ≤ 0.65}, exhibited good performance on the validation set with the area under the receiver operating characteristic curve being 0.984 and balanced accuracy being 0.941 and identified 90 TSHR agonist classes that could not be found previously. The classifier together with the AD_SAL{ρ_s, I_A} may serve as efficient tools for screening EDCs, and the AD characterization methodology may be applied to other ML models.

A consensual machine-learning-assisted QSAR model for effective bioactivity prediction of xanthine oxidase inhibitors using molecular fingerprints

Xanthine oxidase inhibitors (XOIs) have been widely studied due to the promising potential as safe and effective therapeutics in hyperuricemia and gout. Currently, available XOI molecules have been developed from different experiments but they are with the wide structure diversity and significant varying bioactivities. So it is of great practical significance to present a consensual QSAR model for effective bioactivity prediction of XOIs based on a systematic compiling of these XOIs across different experiments. In this work, 249 XOIs belonging to 16 scaffolds were collected and were integrated into a consensual dataset by introducing the concept of IC₅₀ values relative to allopurinol (RIC₅₀). Here, extended connectivity fingerprints (ECFPs) were employed to represent XOI molecules. By performing effective feature selection by machine-learning method, 54 crucial fingerprints were indicated to be valuable for predicting the inhibitory potency (IP) of XOIs. The optimal predictor yields the promising performance by different cross-validation tests. Besides, an external validation of 43 XOIs and a case study on febuxostat also provide satisfactory results, indicating the powerful generalization of our predictor. Here, the predictor was interpreted by shapely additive explanation (SHAP) method which revealed several important substructures by mapping the featured fingerprints to molecular structures. Then, 15 new molecules were designed and predicted by our predictor to show superior IP than febuxostat. Finally, molecular docking simulation was performed to gain a deep insight into molecular binding mode with xanthine oxidase (XO) enzyme, showing that molecules with selenazole moiety, cyano group and isopropyl group tended to yield higher IP. The absorption, distribution, metabolism, excretion and toxicity (ADMET) prediction results further enhanced the potential of these novel XOIs as drug candidates. Overall, this work presents a QSAR model for accurate prediction of IP of XOIs, and is expected to provide new insights for further structure-guided design of novel XOIs.

Genotoxicity of organic contaminants in the soil: A review based on bibliometric analysis and methodological progress

Organic contaminants (OCs) are ubiquitous in the environment, posing severe threats to human health and ecological balance. In particular, OCs and their metabolites could interact with genetic materials to induce genotoxicity, which has attracted considerable attention. In this review, bibliometric analysis was executed to analyze the publications on the genotoxicity of OCs in soil from 1992 to 2021. The result indicated that significant contributions were made by China and the United States in this field and the research hotspots were biological risks, damage mechanisms, and testing methods. Based on this, in this review, we summarized the manifestations and influencing factors of genotoxicity of OCs to soil organisms, the main damage mechanisms, and the most commonly utilized testing methods. OCs can induce genotoxicity and the hierarchical response of soil organisms, which could be influenced by the physicochemical properties of OCs and the properties of soil. Specific mechanisms of genotoxicity can be classified into DNA damage, epigenetic toxicity, and chromosomal aberrations. OCs with different molecular weights lead to genetic material damage by inducing the generation of ROS or forming adducts with DNA, respectively. The micronucleus test and the comet test are the most commonly used testing methods. Moreover, this review also pointed out that future studies should focus on the relationships between bioaccessibilities and genotoxicities, transcriptional regulatory factors, and potential metabolites of OCs to elaborate on the biological risks and mechanisms of genotoxicity from an overall perspective.

Computational Prediction of Drug Phenotypic Effects Based on Substructure-Phenotype Associations

Identifying drug phenotypic effects, including therapeutic effects and adverse drug reactions (ADRs), is an inseparable part for evaluating the potentiality of new drug candidates (NDCs). However, current computational methods for predicting phenotypic effects of NDCs are mainly based on the overall structure of an NDC or a related target. These approaches often lead to inconsistencies between the structures and functions and limit the prediction space of NDCs. In this study, first, we constructed quantitative associations of substructure-domain, domain-ADR, and domain-ATC (Anatomical Therapeutic Chemical Classification System code) through L1LOG and L1SVM machine learning models. These associations represent relationships between phenotypes (ADRs and ATCs) and local structures of drugs and proteins. Then, based on these established associations, substructure-phenotype relationships were constructed which were utilized to quantify drug-phenotype relationships. Thus, this approach could achieve high-throughput and effective evaluations of the druggability of NDCs by referring to the established substructure-phenotype relationships and structural information of NDCs without additional prior knowledge. Using this computational pipeline, 83,205 drug-ATC relationships (including 1,479 drugs and 178 ATCs) and 306,421 drug-ADR relationships (including 1,752 drugs and 454 ADRs) were predicted in total. The prediction results were validated at four levels: five-fold cross validation, public databases, literature, and molecular docking. Furthermore, three case studies demonstrated the feasibility of our method. 79 ATCs and 269 ADRs were predicted to be related to Maraviroc, an approved drug, including the existing antiviral effect in clinical use. Additionally, we also found risk substructures of severe ADRs, for example, SUB215 (>= 1, saturated or only aromatic carbon ring size 7) can result in shock. And we analyzed the mechanism of action (MOA) of interested drugs based on the established drug-substructure-domain-protein associations. In a word, this approach through establishing drug-substructure-phenotype relationships can achieve quantitative prediction of phenotypes for a given NDC or drug without any prior knowledge except its structure information. Using that way, we can directly obtain the relationships between substructure and phenotype of a compound, which is more convenient to analyze the phenotypic mechanism of drugs and accelerate the process of rational drug design.

Machine Learning and Computational Chemistry for the Endocannabinoid System

Computational methods in medicinal chemistry facilitate drug discovery and design. In particular, machine learning methodologies have recently gained increasing attention. This chapter provides a structured overview of the current state of computational chemistry and its applications for the interrogation of the endocannabinoid system (ECS), highlighting methods in structure-based drug design, virtual screening, ligand-based quantitative structure-activity relationship (QSAR) modeling, and de novo molecular design. We emphasize emerging methods in machine learning and anticipate a forecast of future opportunities of computational medicinal chemistry for the ECS.

Identification of molecular initiating events and key events leading to endocrine disrupting effects of PFOA: Integrated molecular dynamic, transcriptomic, and proteomic analyses

Perfluorooctanoic acid (PFOA) can rapidly activate signaling pathways independent of nuclear hormone receptors through membrane receptor regulation, which leads to endocrine disrupting effects. In the present work, the molecular initiating event (MIE) and the key events (KEs) which cause the endocrine disrupting effects of PFOA have been explored and determined based on molecular dynamics simulation (MD), fluorescence analysis, transcriptomics, and proteomics. MD modeling and fluorescence analysis proved that, on binding to the G protein-coupled estrogen receptor-1 (GPER), PFOA could induce a conformational change in the receptor, turning it into an active state. The results also indicated that the binding to GPER was the MIE that led to the adverse outcome (AO) of PFOA. In addition, the downstream signal transduction pathways of GPER, as regulated by PFOA, were further investigated through genomics and proteomics to identify the KEs leading to thr endocrine disrupting effects. Two pathways (Endocrine resistance, ERP and Estrogen signaling pathway, ESP) containing GPER were regulated by different concentration of PFOA and identified as the KEs. The knowledge of MIE, KEs, and AO of PFOA is necessary to understand the links between PFOA and the possible pathways that lead to its negative effects.

Development, Validation, and Application of a Human Reproductive Toxicity Prediction Model Based on Adverse Outcome Pathway

A growing number of environmental contaminants have been proved to have reproductive toxicity to males and females. However, the unclear toxicological mechanism of reproductive toxicants limits the development of virtual screening methods. By consolidating androgen (AR)-/estrogen receptors (ERs)-mediated adverse outcome pathways (AOPs) with more than 8000 chemical substances, we uncovered relationships between chemical features, a series of pathway-related effects, and reproductive apical outcomes─changes in sex organ weights. An AOP-based computational model named RepTox was developed and evaluated to predict and characterize chemicals' reproductive toxicity for males and females. Results showed that RepTox has three outstanding advantages. (I) Compared with the traditional models (37 and 81% accuracy, respectively), AOP significantly improved the predictive robustness of RepTox (96.3% accuracy). (II) Compared with the application domain (AD) of models based on small in vivo datasets, AOP expanded the ADs of RepTox by 1.65-fold for male and 3.77-fold for female, respectively. (III) RepTox implied that hydrophobicity, cyclopentanol substructure, and several topological indices (e.g., hydrogen-bond acceptors) were important, unbiased features associated with reproductive toxicants. Finally, RepTox was applied to the inventory of existing chemical substances of China and identified 2100 and 7281 potential toxicants to the male and female reproductive systems, respectively.

Critical features identification for chemical chronic toxicity based on mechanistic forecast models

Facing billions of tons of pollutants entering the ocean each year, aquatic toxicity is becoming a crucial endpoint for evaluating chemical adverse effects on ecosystems. Notably, huge amount of toxic chemicals at environmental relevant doses can cause potential adverse effects. However, chronic aquatic toxicity effects of chemicals are much scarcer, especially at population level. Rotifers are highly sensitive to toxicants even at chronic low-doses and their communities are usually considered as effective indicators for assessing the status of aquatic ecosystems. Therefore, the no observed effect concentration (NOEC) for population abundance of rotifers were selected as endpoints to develop machine learning models for the prediction of chemical aquatic chronic toxicity. In this study, forty-eight binary models were built by eight types of chemical descriptors combined with six machine learning algorithms. The best binary model was 1D & 2D molecular descriptors - random trees model (RT) with high balanced accuracy (BA) (0.83 for training and 0.83 for validation set), and Matthews correlation coefficient (MCC) (0.72 for training set and 0.67 for validation set). Moreover, the optimal model identified the primary factors (SpMAD_Dzp, AMW, MATS2v) and filtered out three high alerting substructures [c1cc(Cl)cc1, CNCO, CCOP(=S)(OCC)O] influencing the chronic aquatic toxicity. These results showed that the compounds with low molecular volume, high polarity and molecular weight could contribute to adverse effects on rotifers, facilitating the deeper understanding of chronic toxicity mechanisms. In addition, forecast models had better performances than the common models embedded into ECOSAR software. This study provided insights into structural features responsible for the toxicity of different groups of chemicals and thereby allowed for the rational design of green and safer alternatives.

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.

MegaSyn: Integrating Generative Molecular Design, Automated Analog Designer, and Synthetic Viability Prediction

Generative machine learning models have become widely adopted in drug discovery and other fields to produce new molecules and explore molecular space, with the goal of discovering novel compounds with optimized properties. These generative models are frequently combined with transfer learning or scoring of the physicochemical properties to steer generative design, yet often, they are not capable of addressing a wide variety of potential problems, as well as converge into similar molecular space when combined with a scoring function for the desired properties. In addition, these generated compounds may not be synthetically feasible, reducing their capabilities and limiting their usefulness in real-world scenarios. Here, we introduce a suite of automated tools called MegaSyn representing three components: a new hill-climb algorithm, which makes use of SMILES-based recurrent neural network (RNN) generative models, analog generation software, and retrosynthetic analysis coupled with fragment analysis to score molecules for their synthetic feasibility. We show that by deconstructing the targeted molecules and focusing on substructures, combined with an ensemble of generative models, MegaSyn generally performs well for the specific tasks of generating new scaffolds as well as targeted analogs, which are likely synthesizable and druglike. We now describe the development, benchmarking, and testing of this suite of tools and propose how they might be used to optimize molecules or prioritize promising lead compounds using these RNN examples provided by multiple test case examples.

Graph Attention Network Model with Defined Applicability Domains for Screening PBT Chemicals

In silico models for screening environmentally persistent, bio-accumulative, and toxic (PBT) substances are necessary for sound management of chemicals. Due to the complex structure-activity landscapes (SALs) on the PBT attributes, previous models for screening PBT chemicals lack either applicability domain (AD) characterizations or interpretability, restricting their applications. Herein, graph attention networks (GATs), a novel neural network architecture, were introduced to construct models for screening PBT chemicals. Results show that the GAT model not only outperformed those in previous studies but also exhibited interpretability since it optimizes attention weight parameters (P_AW) that indicate contributions of each atom to the PBT attributes. An AD characterization termed AD_FP-AC, which considers both molecular fingerprint (FP) similarities and compounds at activity cliffs (ACs) of SALs, was proposed to describe the ADs, which further assured the performance of the GAT model. Eight previously unidentified classes of compounds were identified as PBT chemicals from the Inventory of Existing Chemical Substances in China. The GAT model together with the AD_FP-AC characterization may serve as efficient tools for screening PBT chemicals, and the modeling methodology can be applied to other physicochemical, environmental, behavioral, and toxicological parameters of chemicals that are necessary for their risk assessment and management.

Machine learning models on chemical inhibitors of mitochondrial electron transport chain

Chemicals can induce adverse effects in humans by inhibiting mitochondrial electron transport chain (ETC) such as disrupting mitochondrial membrane potential, enhancing oxidative stress and causing some diseases. Thus, identifying ETC inhibitors (ETCi) is important to chemical risk assessment and protecting the public health. However, it is not feasible to identify all ETCi with experimental methods. Quantitative structure-activity relationship (QSAR) modeling is a promising method to rapidly and effectively identify ETCi. In this study, QSAR models for predicting ETCi were developed using machine learning methods. A clustering-based under-sampling (CBUS) method was developed to handle the imbalance issue in training sets. Structure-activity landscapes were generated and analyzed for training sets generated by the CBUS method. The consensus QSAR models constructed with CBUS achieved satisfactory performances (balanced accuracy = 0.852) in 100 iterations of five-fold cross validations, indicating the models can effectively classify ETCi. The classification model was further employed to screen chemicals in the Inventory of Existing Chemical Substances of China and 13 chemicals were identified as ETCi. Fifteen structural alerts for ETCi were identified in this study. These results demonstrated that the model and structural alerts are useful to screen ETCi.

Potential Application of Machine-Learning-Based Quantum Chemical Methods in Environmental Chemistry

It is an important topic in environmental sciences to understand the behavior and toxicology of chemical pollutants. Quantum chemical methodologies have served as useful tools for probing behavior and toxicology of chemical pollutants in recent decades. In recent years, machine learning (ML) techniques have brought revolutionary developments to the field of quantum chemistry, which may be beneficial for investigating environmental behavior and toxicology of chemical pollutants. However, the ML-based quantum chemical methods (ML-QCMs) have only scarcely been used in environmental chemical studies so far. To promote applications of the promising methods, this Perspective summarizes recent progress in the ML-QCMs and focuses on their potential applications in environmental chemical studies that could hardly be achieved by the conventional quantum chemical methods. Potential applications and challenges of the ML-QCMs in predicting degradation networks of chemical pollutants, searching global minima for atmospheric nanoclusters, discovering heterogeneous or photochemical transformation pathways of pollutants, as well as predicting environmentally relevant end points with wave functions as descriptors are introduced and discussed.

Machine Learning Models for Predicting Cytotoxicity of Nanomaterials

The wide application of nanomaterials in consumer and medical products has raised concerns about their potential adverse effects on human health. Thus, more and more biological assessments regarding the toxicity of nanomaterials have been performed. However, the different ways the evaluations were performed, such as the utilized assays, cell lines, and the differences of the produced nanoparticles, make it difficult for scientists to analyze and effectively compare toxicities of nanomaterials. Fortunately, machine learning has emerged as a powerful tool for the prediction of nanotoxicity based on the available data. Among different types of toxicity assessments, nanomaterial cytotoxicity was the focus here because of the high sensitivity of cytotoxicity assessment to different treatments without the need for complicated and time-consuming procedures. In this review, we summarized recent studies that focused on the development of machine learning models for prediction of cytotoxicity of nanomaterials. The goal was to provide insight into predicting potential nanomaterial toxicity and promoting the development of safe nanomaterials.

Carcinogenic Risk of 2,6-Di-tert-Butylphenol and Its Quinone Metabolite 2,6-DTBQ Through Their Interruption of RARβ: In Vivo, In Vitro, and In Silico Investigations

Thousands of contaminants are used worldwide and eventually released into the environment, presenting a challenge of health risk assessment. The identification of key toxic pathways and characterization of interactions with target biomacromolecules are essential for health risk assessments. The adverse outcome pathway (AOP) incorporates toxic mechanisms into health risk assessment by emphasizing the relationship among molecular initiating events (MIEs), key events (KEs), and adverse outcome (AO). Herein, we attempted the use of AOP to decipher the toxic effects of 2,6-di-tert-butylphenol (2,6-DTBP) and its para-quinone metabolite 2,6-di-tert-butyl-1,4-benzoquinone (2,6-DTBQ) based on integrated transcriptomics, molecular modeling, and cell-based assays. Through transcriptomics and quantitative real-time PCR validation, we identified retinoic acid receptor β (RARβ) as the key target biomacromolecule. The epigenetic analysis and molecular modeling revealed RARβ interference as one MIE, including DNA methylation and conformational changes. In vitro assays extended subsequent KEs, including altered protein expression of p-Erk1/2 and COX-2, and promoted cancer cell H4IIE proliferation and metastasis. These toxic effects altogether led to carcinogenic risk as the AO of 2,6-DTBP and 2,6-DTBQ, in line with chemical carcinogenesis identified from transcriptome profiling. Overall, our simplified AOP network of 2,6-DTBP and 2,6-DTBQ facilitates relevant health risk assessment.

A QSAR-ICE-SSD Model Prediction of the PNECs for Per- and Polyfluoroalkyl Substances and Their Ecological Risks in an Area of Electroplating Factories

Per- and polyfluoroalkyl substances (PFASs) are a class of highly fluorinated aliphatic compounds that are persistent and bioaccumulate, posing a potential threat to the aquatic environment. The electroplating industry is considered to be an important source of PFASs. Due to emerging PFASs and many alternatives, the acute toxicity data for PFASs and their alternatives are relatively limited. In this study, a QSAR–ICE–SSD composite model was constructed by combining quantitative structure-activity relationship (QSAR), interspecies correlation estimation (ICE), and species sensitivity distribution (SSD) models in order to obtain the predicted no-effect concentrations (PNECs) of selected PFASs. The PNECs for the selected PFASs ranged from 0.254 to 6.27 mg/L. The ΣPFAS concentrations ranged from 177 to 983 ng/L in a river close to an electroplating industry in Shenzhen. The ecological risks associated with PFASs in the river were below 2.97 × 10⁻⁴.

Machine learning in natural and engineered water systems

Water resources of desired quality and quantity are the foundation for human survival and sustainable development. To better protect the water environment and conserve water resources, efficient water management, purification, and transportation are of critical importance. In recent years, machine learning (ML) has exhibited its practicability, reliability, and high efficiency in numerous applications; furthermore, it has solved conventional and emerging problems in both natural and engineered water systems. For example, ML can predict various water quality indicators in situ and real-time by considering the complex interactions among water-related variables. ML approaches can also solve emerging pollution problems with proven rules or universal mechanisms summarized from the related research. Moreover, by applying image recognition technology to analyze the relationships between image information and physicochemical properties of the research object, ML can effectively identify and characterize specific contaminants. In view of the bright prospects of ML, this review comprehensively summarizes the development of ML applications in natural and engineered water systems. First, the concept and modeling steps of ML are briefly introduced, including data preparation, algorithm selection and model evaluation. In addition, comprehensive applications of ML in recent studies, including predicting water quality, mapping groundwater contaminants, classifying water resources, tracing contaminant sources, and evaluating pollutant toxicity in natural water systems, as well as modeling treatment techniques, assisting characterization analysis, purifying and distributing drinking water, and collecting and treating sewage water in engineered water systems, are summarized. Finally, the advantages and disadvantages of commonly used algorithms are analyzed according to their structures and mechanisms, and recommendations on the selection of ML algorithms for different studies, as well as prospects on the application and development of ML in water science are proposed. This review provides references for solving a wider range of water-related problems and brings further insights into the intelligent development of water science.