Integrating in silico models to enhance predictivity for developmental toxicity

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 summarize 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 approach, we expound the research status, existing challenges, and feasible solutions. Finally, several new directions and suggestions for toxicity prediction are also put forward.

Where developmental toxicity meets explainable artificial intelligence: state-of-the-art and perspectives

INTRODUCTION

The application of Artificial Intelligence (AI) to predictive toxicology is rapidly increasing, particularly aiming to develop non-testing methods that effectively address ethical concerns and reduce economic costs. In this context, Developmental Toxicity (Dev Tox) stands as a key human health endpoint, especially significant for safeguarding maternal and child well-being.

AREAS COVERED

This review outlines the existing methods employed in Dev Tox predictions and underscores the benefits of utilizing New Approach Methodologies (NAMs), specifically focusing on eXplainable Artificial Intelligence (XAI), which proves highly efficient in constructing reliable and transparent models aligned with recommendations from international regulatory bodies.

EXPERT OPINION

The limited availability of high-quality data and the absence of dependable Dev Tox methodologies render XAI an appealing avenue for systematically developing interpretable and transparent models, which hold immense potential for both scientific evaluations and regulatory decision-making.

MCPNET: Development of an interpretable deep learning model based on multiple conformations of the compound for predicting developmental toxicity

The development of deep learning models for predicting toxicological endpoints has shown great promise, but one of the challenges in the field is the accuracy and interpretability of these models. The bioactive conformation of a compound plays a critical role for it to bind in the target. It is a big issue to figure out the bioactive conformation in deep learning without the co-crystal structure or highly precise molecular simulations. In this study, we developed a deep learning framework of Multi-Conformation Point Network (MCPNET) to construct classification and regression models, respectively, based on electrostatic potential distributions on vdW surfaces around multiple conformations of the compound using a dataset of compounds with developmental toxicity in zebrafish embryo. MCPNET applied 3D multi-conformational surface point cloud to extract the molecular features for model training, which may be critical for capturing the structural diversity of compounds. The models achieved an accuracy of 85 % on the classification task and R2 of 0.66 on the regression task, outperforming traditional machine learning models and other deep learning models. The key feature of our model is its interpretability with the component visualization to identify the factors contributing to the prediction and to understand the compound action mechanism. MCPNET may predict the conformation quietly close to the bioactive conformation of a compound by attention-based multi-conformation pooling mechanism. Our results demonstrated the potential of deep learning based on 3D molecular representations in accurately predicting developmental toxicity. The source code is publicly available at https://github.com/Superlit-CC/MCPNET.

TISBE: A Public Web Platform for the Consensus-Based Explainable Prediction of Developmental Toxicity

Despite being extremely relevant for the protection of prenatal and neonatal health, the developmental toxicity (Dev Tox) is a highly complex endpoint whose molecular rationale is still largely unknown. The lack of availability of high-quality data as well as robust nontesting methods makes its understanding even more difficult. Thus, the application of new explainable alternative methods is of utmost importance, with Dev Tox being one of the most animal-intensive research themes of regulatory toxicology. Descending from TIRESIA (Toxicology Intelligence and Regulatory Evaluations for Scientific and Industry Applications), the present work describes TISBE (TIRESIA Improved on Structure-Based Explainability), a new public web platform implementing four fundamental advancements for in silico analyses: a three times larger dataset, a transparent XAI (explainable artificial intelligence) framework employing a fragment-based fingerprint coding, a novel consensus classifier based on five independent machine learning models, and a new applicability domain (AD) method based on a double top-down approach for better estimating the prediction reliability. The training set (TS) includes as many as 1008 chemicals annotated with experimental toxicity values. Based on a 5-fold cross-validation, a median value of 0.410 for the Matthews correlation coefficient was calculated; TISBE was very effective, with a median value of sensitivity and specificity equal to 0.984 and 0.274, respectively. TISBE was applied on two external pools made of 1484 bioactive compounds and 85 pediatric drugs taken from ChEMBL (Chemical European Molecular Biology Laboratory) and TEDDY (Task-Force in Europe for Drug Development in the Young) repositories, respectively. Notably, TISBE gives users the option to clearly spot the molecular fragments responsible for the toxicity or the safety of a given chemical query and is available for free at https://prometheus.farmacia.uniba.it/tisbe.

Making sense of chemical space network shows signs of criticality

Chemical space modelling has great importance in unveiling and visualising latent information, which is critical in predictive toxicology related to drug discovery process. While the use of traditional molecular descriptors and fingerprints may suffer from the so-called curse of dimensionality, complex networks are devoid of the typical drawbacks of coordinate-based representations. Herein, we use chemical space networks (CSNs) to analyse the case of the developmental toxicity (Dev Tox), which remains a challenging endpoint for the difficulty of gathering enough reliable data despite very important for the protection of the maternal and child health. Our study proved that the Dev Tox CSN has a complex non-random organisation and can thus provide a wealth of meaningful information also for predictive purposes. At a phase transition, chemical similarities highlight well-established toxicophores, such as aryl derivatives, mostly neurotoxic hydantoins, barbiturates and amino alcohols, steroids, and volatile organic compounds ether-like chemicals, which are strongly suspected of the Dev Tox onset and can thus be employed as effective alerts for prioritising chemicals before testing.

Advancing the use of new approach methodologies for assessing teratogenicity: Building a tiered approach

Many New Approach Methodologies (NAMs) have been developed for the safety assessment of new ingredients. Research into reproductive toxicity and teratogenicity is a particularly high priority, especially given their mechanistic complexity. Forty-six non-teratogenic and 39 teratogenic chemicals were screened for teratogenic potential using the in silico DART model from the OECD QSAR Toolbox; the devTox quickPredict™ (devTox assay) test and the Zebrafish Embryotoxicity Test (ZET). The sensitivity and specificity were 94.7% and 84.1%, respectively, for the DART tree (83 chemicals), 86.1% and 35.6% for the devTox (81 chemicals) and 77.8% and 76.7% for the ZET (57 chemicals). Fifty-three chemicals were tested in all three assays and when results were combined and based on a "2 out of 3 rule", the sensitivity and specificity were 96.0% and 71.4%, respectively. The specificity of the devTox assay for a sub-set of 43 chemicals was increased from 26.1% to 82.6% by incorporating human plasma concentrations into the assay interpretation. When all 85 chemicals were assessed in a decision tree approach, there was an excellent predictivity and assay robustness of 90%. In conclusion, all three models exhibited a good sensitivity and specificity, especially when outcomes from all three were combined or used in "2 out of 3" or a tiered decision tree approach. The latter is an interesting predictive approach for evaluating the teratogenic potential of new chemicals. Future investigations will extend the number of chemicals tested, as well as explore ways to refine the results and obtain a robust Integrated Testing Strategy to evaluate teratogenic potential.

Guidance document on the impact of water treatment processes on residues of active substances or their metabolites in water abstracted for the production of drinking water

This guidance document provides a tiered framework for risk assessors and facilitates risk managers in making decisions concerning the approval of active substances (AS) that are chemicals in plant protection products (PPPs) and biocidal products, and authorisation of the products. Based on the approaches presented in this document, a conclusion can be drawn on the impact of water treatment processes on residues of the AS or its metabolites in surface water and/or groundwater abstracted for the production of drinking water, i.e. the formation of transformation products (TPs). This guidance enables the identification of actual public health concerns from exposure to harmful compounds generated during the processing of water for the production of drinking water, and it focuses on water treatment methods commonly used in the European Union (EU). The tiered framework determines whether residues from PPP use or residues from biocidal product use can be present in water at water abstraction locations. Approaches, including experimental methods, are described that can be used to assess whether harmful TPs may form during water treatment and, if so, how to assess the impact of exposure to these water treatment TPs (tTPs) and other residues including environmental TPs (eTPs) on human and domesticated animal health through the consumption of TPs via drinking water. The types of studies or information that would be required are described while avoiding vertebrate testing as much as possible. The framework integrates the use of weight-of-evidence and, when possible alternative (new approach) methods to avoid as far as possible the need for additional testing.

TIRESIA: An eXplainable Artificial Intelligence Platform for Predicting Developmental Toxicity

Herein, a robust and reproducible eXplainable Artificial Intelligence (XAI) approach is presented, which allows prediction of developmental toxicity, a challenging human-health endpoint in toxicology. The application of XAI as an alternative method is of the utmost importance with developmental toxicity being one of the most animal-intensive areas of regulatory toxicology. In this work, the established CAESAR (Computer Assisted Evaluation of industrial chemical Substances According to Regulations) training set made of 234 chemicals for model learning is employed. Two test sets, including as a whole 585 chemicals, were instead used for validation and generalization purposes. The proposed framework favorably compares with the state-of-the-art approaches in terms of accuracy, sensitivity, and specificity, thus resulting in a reliable support system for developmental toxicity ensuring informativeness, uncertainty estimation, generalization, and transparency. Based on the eXtreme Gradient Boosting (XGB) algorithm, our predictive model provides easy interpretative keys based on specific molecular descriptors and structural alerts enabling one to distinguish toxic and nontoxic chemicals. Inspired by the Organisation for Economic Co-operation and Development (OECD) principles for the validation of Quantitative Structure-Activity Relationships (QSARs) for regulatory purposes, the results are summarized in a standard report in portable document format, enclosing also details concerned with a density-based model applicability domain and SHAP (SHapley Additive exPlanations) explainability, the latter particularly useful to better understand the effective roles played by molecular features. Notably, our model has been implemented in TIRESIA (Toxicology Intelligence and Regulatory Evaluations for Scientific and Industry Applications), a free of charge web platform available at http://tiresia.uniba.it.

Human-Based New Approach Methodologies in Developmental Toxicity Testing: A Step Ahead from the State of the Art with a Feto-Placental Organ-on-Chip Platform

Developmental toxicity testing urgently requires the implementation of human-relevant new approach methodologies (NAMs) that better recapitulate the peculiar nature of human physiology during pregnancy, especially the placenta and the maternal/fetal interface, which represent a key stage for human lifelong health. Fit-for-purpose NAMs for the placental-fetal interface are desirable to improve the biological knowledge of environmental exposure at the molecular level and to reduce the high cost, time and ethical impact of animal studies. This article reviews the state of the art on the available in vitro (placental, fetal and amniotic cell-based systems) and in silico NAMs of human relevance for developmental toxicity testing purposes; in addition, we considered available Adverse Outcome Pathways related to developmental toxicity. The OECD TG 414 for the identification and assessment of deleterious effects of prenatal exposure to chemicals on developing organisms will be discussed to delineate the regulatory context and to better debate what is missing and needed in the context of the Developmental Origins of Health and Disease hypothesis to significantly improve this sector. Starting from this analysis, the development of a novel human feto-placental organ-on-chip platform will be introduced as an innovative future alternative tool for developmental toxicity testing, considering possible implementation and validation strategies to overcome the limitation of the current animal studies and NAMs available in regulatory toxicology and in the biomedical field.

Molecular Characterization of Estrogen Receptor Agonists during Sewage Treatment Processes Using Effect-Directed Analysis Combined with High-Resolution Full-Scan Screening

Endocrine-disrupting potential was evaluated during the sewage treatment process using in vitro bioassays. Aryl hydrocarbon receptor (AhR)-, androgen receptor (AR)-, glucocorticoid receptor (GR)-, and estrogen receptor (ER)-mediated activities were assessed over five steps of the treatment process. Bioassays of organic extracts showed that AhR, AR, and GR potencies tended to decrease through the sewage treatment process, whereas ER potencies did not significantly decrease. Bioassays on reverse-phase high-performance liquid chromatography fractions showed that F5 (log K_OW 2.5-3.0) had great ER potencies. Full-scan screening of these fractions detected two novel ER agonists, arenobufagin and loratadine, which are used pharmaceuticals. These compounds accounted for 3.3-25% of the total ER potencies and 4% of the ER potencies in the final effluent. The well-known ER agonists, estrone and 17β-estradiol, accounted for 60 and 17% of the ER potencies in F5 of the influent and primary treatment, respectively. Fourier transform ion cyclotron resonance mass spectrometry analysis showed that various molecules were generated during the treatment process, especially CHO and CHOS (C: carbon, H: hydrogen, O: oxygen, and S: sulfur). This study documented that widely used pharmaceuticals are introduced into the aquatic environments without being removed during the sewage treatment process.

Identification of novel polar aryl hydrocarbon receptor agonists accumulated in liver of black-tailed gulls in Korea using advanced effect-directed analysis

Although bioaccumulation of persistent organic pollutants in seabirds has been examined, few studies have been conducted to identify previously unidentified substances. Here, aryl hydrocarbon receptor (AhR) agonists were identified in livers of black-tailed gulls from South Korea using effect-directed analysis combined with full-scan screening analysis. Significant AhR-mediated potencies were observed in the polar fractions of liver extracts using H4IIE-luc bioassay. Eight known polar AhR agonists accounted for 11-20% of the total AhR-mediated potencies in the polar fractions; hydrocortisone and rutaecarpine were the major contributors. Twenty-two AhR agonist candidates in the polar fractions were identified using liquid chromatography-quadrupole time-of-flight mass spectrometry during a six-step selection process. Of these, [10]-gingerol, angelicin, corticosterone, eupatilin, etofenprox, oxadixyl, and tretinoin were identified as novel AhR agonists. The contribution to potencies increased with inclusion of novel AhR agonists (27-52%); corticosterone and [10]-gingerol contributed significantly. Quantitative structure-activity relationship suggested that the novel AhR agonists have other potential toxic effects, including carcinogenicity and mutagenicity. Polar AhR agonists have been used for pharmaceuticals and pesticides. Some novel AhR agonists have log K_OW > 2 and log K_OA ≥ 6, which indicates that these compounds can be biomagnified in air-breathing organisms, such as seabirds.

Optimization of the TeraTox assay for preclinical teratogenicity assessment

Current animal-free methods to assess teratogenicity of drugs under development still deliver high numbers of false negatives. To improve the sensitivity of human teratogenicity prediction, we characterized the TeraTox test, a newly developed multilineage differentiation assay using 3D human-induced pluripotent stem cells. TeraTox produces primary output concentration-dependent cytotoxicity and altered gene expression induced by each test compound. These data are fed into an interpretable machine-learning model to perform prediction, which relates to the concentration-dependent human teratogenicity potential of drug candidates. We applied TeraTox to profile 33 approved pharmaceuticals and 12 proprietary drug candidates with known in vivo data. Comparing TeraTox predictions with known human or animal toxicity, we report an accuracy of 69% (specificity: 53%, sensitivity: 79%). TeraTox performed better than 2 quantitative structure-activity relationship models and had a higher sensitivity than the murine embryonic stem cell test (accuracy: 58%, specificity: 76%, and sensitivity: 46%) run in the same laboratory. The overall prediction accuracy could be further improved by combining TeraTox and mouse embryonic stem cell test results. Furthermore, patterns of altered gene expression revealed by TeraTox may help grouping toxicologically similar compounds and possibly deducing common modes of action. The TeraTox assay and the dataset described here therefore represent a new tool and a valuable resource for drug teratogenicity assessment.

Effect-directed identification of novel aryl hydrocarbon receptor-active aromatic compounds in coastal sediments collected from a highly industrialized area

In this study, we identified major aryl hydrocarbon receptor (AhR) agonists in the sediments from Yeongil Bay (n = 6) using effect-directed analysis. Using the H4IIE-luc bioassays, great AhR-mediated potencies were found in aromatic fractions (F2) of sediment organic extracts from silica gel column chromatography and sub-fractions (F2.6-F2.8) from reverse phase-HPLC. Full-scan mass spectrometric analysis using GC-QTOFMS was conducted to identify novel AhR agonists in highly potent fractions, such as F2.6-F2.8 of S1 (Gumu Creek). Selection criteria for AhR-active compounds consisted of three steps, including matching factor of NIST library (≥70), aromatic structures, and the number of aromatic rings (≥4). Fifty-nine compounds were selected as tentative AhR agonist candidates, with the AhR-mediated activity being assessed for six compounds for which standard materials were available commercially. Of these compounds, 20-methylcholanthrene, 7-methylbenz[a]anthracene, 10-methylbenz[a]pyrene, and 7,12-dimethylbenz[a]anthracene exhibited significant AhR-mediated potency. Relative potency values of these compounds were determined relative to benzo[a]pyrene to be 3.2, 1.4, 1.2, and 0.2, respectively. EPA positive matrix factorization modeling indicated that the sedimentary AhR-active aromatic compounds primarily originated from coal combustion and vehicle emissions. Potency balance analysis indicated that four novel AhR agonists explained 0.007% to 1.7% of bioassay-derived AhR-mediated potencies in samples.

Comparison of predictions of developmental toxicity for compounds of solvent data set

We have considered a series of 235 compounds technically classified as solvents. Chemically, they belong to different classes. Their potential developmental toxicity was evaluated using two models available on platform VEGA HUB; model CAESAR and the Developmental/ Reproductive Toxicity library (PG) model. Models provide beside the prediction of developmental toxicity additional information on similar compounds from models' training sets. In the report, first, we compare the predictions. Second, the sets of similar compounds have been used to implement the clustering scheme. The Kohonen artificial neural network method has been applied as a clustering method. The clusters obtained have been discussed for both models.

'In silico' toxicology methods in drug safety assessment

While experimental animal investigation has historically been the most conventional approach conducted to assess drug safety and is currently considered the main method for determining drug toxicity, these studies are constricted by cost, time, and ethical approvals. Over the last 20 years, there have been significant advances in computational sciences and computer data processing, while knowledge of alternative techniques and their application has developed into a valuable skill in toxicology. Thus, the application of in silico methods in drug safety assessment is constantly increasing. They are very complex and are grounded on accumulated knowledge from toxicology, bioinformatics, biochemistry, statistics, mathematics, as well as molecular biology. This review will summarize current state-of-the-art scientific data on the use of in silico methods in toxicity testing, taking into account their shortcomings, and highlighting the strategies that should deliver consistent results, while covering the applications of in silico methods in preclinical trials and drug impurities toxicity testing.

Current contamination status of traditional and emerging persistent toxic substances in the sediments of Ulsan Bay, South Korea

Contamination status of traditional and emerging persistent toxic substances (PTSs) in sediments and their major sources were investigated in Ulsan Bay, Korea. A total of 47 PTSs, including 15 traditional PAHs, ten styrene oligomers (SOs), six alkylphenols (APs), and 16 emerging PAHs (E-PAHs) were analyzed. Concentrations of traditional PAHs, SOs, and APs ranged from 35 to 1300 ng g^-1 dry weight (dw), 30 to 3800 ng g^-1 dw, and 30 to 430 ng g^-1 dw, respectively. For the last 20 years, PTSs contamination in the bay area has been improved. However, 12 E-PAHs were widely detected in sediments, with a maximum of 240 ng g^-1 dw (for benzo[e]pyrene) at the creek site. These E-PAHs seemed to originate from surrounding activities, such as biomass combustion, mobile sources, and diesel combustion. Due to environmental concerns for E-PAHs, further research on the potential toxicity, distribution, and behavior of these compounds should be implemented.

Zebrafish AC50 modelling: (Q)SAR models to predict developmental toxicity in zebrafish embryo

Developmental toxicity refers to the occurrence of adverse effects on a developing organism as a consequence of exposure to hazardous chemicals. The assessment of developmental toxicity has become relevant to the safety assessment process of chemicals. The zebrafish embryo developmental toxicology assay is an emerging test used to screen the teratogenic potential of chemicals and it is proposed as a promising test to replace teratogenic assays with animals. Supported by the increased availability of data from this test, the developmental toxicity assay with zebrafish has become an interesting endpoint for the in silico modelling. The purpose of this study was to build up quantitative structure-activity relationship (QSAR) models. In this work, new in silico models for the evaluation of developmental toxicity were built using a well-defined set of data from the ToxCast^TM Phase I chemical library on the zebrafish embryo. Categorical and continuous QSAR models were built by gradient boosting machine learning and the Monte Carlo technique respectively, in accordance with Organization for Economic Co-operation and Development principles and their statistical quality was satisfactory. The classification model reached balanced accuracy 0.89 and Matthews correlation coefficient 0.77 on the test set. The regression model reached correlation coefficient R² 0.70 in external validation and leave-one-out cross-validated Q² 0.73 in internal validation.

Computational approach for collection and prediction of molecular initiating events in developmental toxicity

Developmental toxicity is defined as the occurrence of adverse effects on the developing organism as a result from exposure to a toxic agent. These alterations can have long-term acute effects. Current in vitro models present important limitations and the evaluation of toxicity is not entirely objective. In silico methods have also shown limited success, in part due to complex and varied mechanisms of action that mediate developmental toxicity, which are sometimes poorly understood. In this article, we compiled a dataset of compounds with developmental toxicity categories and annotated mechanisms of action for both toxic and non-toxic compounds (DVTOX). With it, we selected a panel of protein targets that might be part of putative Molecular Initiating Events (MIEs) of Adverse Outcome Pathways of developmental toxicity. The validity of this list of candidate MIEs was studied through the evaluation of new drug-target relationships that include such proteins, but were not part of the original database. Finally, an orthology analysis of this protein panel was conducted to select an appropriate animal model to assess developmental toxicity. We tested our approach using the zebrafish embryo toxicity test, finding positive results.

Consensus versus Individual QSARs in Classification: Comparison on a Large-Scale Case Study

Consensus strategies have been widely applied in many different scientific fields, based on the assumption that the fusion of several sources of information increases the outcome reliability. Despite the widespread application of consensus approaches, their advantages in quantitative structure–activity relationship (QSAR) modeling have not been thoroughly evaluated, mainly due to the lack of appropriate large-scale data sets. In this study, we evaluated the advantages and drawbacks of consensus approaches compared to single classification QSAR models. To this end, we used a data set of three properties (androgen receptor binding, agonism, and antagonism) for approximately 4000 molecules with predictions performed by more than 20 QSAR models, made available in a large-scale collaborative project. The individual QSAR models were compared with two consensus approaches, majority voting and the Bayes consensus with discrete probability distributions, in both protective and nonprotective forms. Consensus strategies proved to be more accurate and to better cover the analyzed chemical space than individual QSARs on average, thus motivating their widespread application for property prediction. Scripts and data to reproduce the results of this study are available for download.

Predicting the safety of medicines in pregnancy: A workshop report

The framework for developmental toxicity testing has remained largely unchanged for over 50 years and although it remains invaluable in assessing potential risks in pregnancy, knowledge gaps exist, and some outcomes do not necessarily correlate with clinical experience. Advances in omics, in silico approaches and alternative assays are providing opportunities to enhance our understanding of embryo-fetal development and the prediction of potential risks associated with the use of medicines in pregnancy. A workshop organised by the Medicines and Healthcare products Regulatory Agency (MHRA), "Predicting the Safety of Medicines in Pregnancy - a New Era?", was attended by delegates representing regulatory authorities, academia, industry, patients, funding bodies and software developers to consider how to improve the quality of and access to nonclinical developmental toxicity data and how to use this data to better predict the safety of medicines in human pregnancy. The workshop delegates concluded that based on comparative data to date alternative methodologies are currently no more predictive than conventional methods and not qualified for use in regulatory submissions. To advance the development and qualification of alternative methodologies, there is a requirement for better coordinated multidisciplinary cross-sector interactions coupled with data sharing. Furthermore, a better understanding of human developmental biology and the incorporation of this knowledge into the development of alternative methodologies is essential to enhance the prediction of adverse outcomes for human development. The output of the workshop was a series of recommendations aimed at supporting multidisciplinary efforts to develop and validate these alternative methodologies.

Value and limitation of structure-based profilers to characterize developmental and reproductive toxicity potential

The uncertainty regarding the safety of chemicals leaching from food packaging triggers attention. In silico models provide solutions for screening of these chemicals, since many are toxicologically uncharacterized. For hazard assessment, information on developmental and reproductive toxicity (DART) is needed. The possibility to apply in silico toxicology to identify and quantify DART alerts was investigated. Open-source models and profilers were applied to 195 packaging chemicals and analogues. An approach based on DART and estrogen receptor (ER) binding profilers and molecular docking was able to identify all except for one chemical with documented DART properties. Twenty percent of the chemicals in the database known to be negative in experimental studies were classified as positive. The scheme was then applied to 121 untested chemicals. Alerts were identified for sixteen of them, five being packaging substances, the others structural analogues. Read-across was then developed to translate alerts into quantitative toxicological values. They can be used to calculate margins of exposure (MoE), the size of which reflects safety concern. The application of this approach appears valuable for hazard characterization of toxicologically untested packaging migrants. It is an alternative to the use of default uncertainty factor (UF) applied to animal chronic toxicity value to handle absence of DART data in hazard characterization.

Major AhR-active chemicals in sediments of Lake Sihwa, South Korea: Application of effect-directed analysis combined with full-scan screening analysis

This study utilized effect-directed analysis (EDA) combined with full-scan screening analysis (FSA) to identify aryl hydrocarbon receptor (AhR)-active compounds in sediments of inland creeks flowing into Lake Sihwa, South Korea. The specific objectives were to (i) investigate the major AhR-active fractions of organic extracts of sediments by using H4IIE-luc in vitro bioassay (4 h and 72 h exposures), (ii) quantify known AhR agonists, such as polycyclic aromatic hydrocarbons (PAHs) and styrene oligomers (SOs), (iii) identify unknown AhR agonists by use of gas chromatography-quadrupole time-of-flight mass spectrometry (GC-QTOFMS), and (iv) determine contributions of AhR agonists to total potencies measured by use of the bioassay. FSA was conducted on fractions F2.6 and F2.7 (aromatics with log K_ow 5-7) in extracts of sediment from Siheung Creek (industrial area). Those fractions exhibited significant AhR-mediated potency as well as relatively great concentrations of PAHs and SOs. FSA detected 461 and 449 compounds in F2.6 and F2.7, respectively. Of these, five tentative candidates of AhR agonist were selected based on NIST library matching, aromatic structures and numbers of rings, and available standards. Benz[b]anthracene, 11H-benzo[a]fluorene, and 4,5-methanochrysene exhibited significant AhR-mediated potency in the H4IIE-luc bioassay, and relative potencies of these compounds were determined. Potency balance analysis demonstrated that these three newly identified AhR agonists explained 1.1% to 67% of total induced AhR-mediated potencies of samples, which were particularly great for industrial sediments. Follow-up studies on sources and ecotoxicological effects of these compounds in coastal environments would be required.

Integrating in silico models and read-across methods for predicting toxicity of chemicals: A step-wise strategy

In silico methods and models are increasingly used for predicting properties of chemicals for hazard identification and hazard characterisation in the absence of experimental toxicity data. Many in silico models are available and can be used individually or in an integrated fashion. Whilst such models offer major benefits to toxicologists, risk assessors and the global scientific community, the lack of a consistent framework for the integration of in silico results can lead to uncertainty and even contradictions across models and users, even for the same chemicals. In this context, a range of methods for integrating in silico results have been proposed on a statistical or case-specific basis. Read-across constitutes another strategy for deriving reference points or points of departure for hazard characterisation of untested chemicals, from the available experimental data for structurally-similar compounds, mostly using expert judgment. Recently a number of software systems have been developed to support experts in this task providing a formalised and structured procedure. Such a procedure could also facilitate further integration of the results generated from in silico models and read-across. This article discusses a framework on weight of evidence published by EFSA to identify the stepwise approach for systematic integration of results or values obtained from these "non-testing methods". Key criteria and best practices for selecting and evaluating individual in silico models are also described, together with the means to combining the results, taking into account any limitations, and identifying strategies that are likely to provide consistent results.

Prediction Model with High-Performance Constitutive Androstane Receptor (CAR) Using DeepSnap-Deep Learning Approach from the Tox21 10K Compound Library

The constitutive androstane receptor (CAR) plays pivotal roles in drug-induced liver injury through the transcriptional regulation of drug-metabolizing enzymes and transporters. Thus, identifying regulatory factors for CAR activation is important for understanding its mechanisms. Numerous studies conducted previously on CAR activation and its toxicity focused on in vivo or in vitro analyses, which are expensive, time consuming, and require many animals. We developed a computational model that predicts agonists for the CAR using the Toxicology in the 21st Century 10k library. Additionally, we evaluate the prediction performance of novel deep learning (DL)-based quantitative structure-activity relationship analysis called the DeepSnap-DL approach, which is a procedure of generating an omnidirectional snapshot portraying three-dimensional (3D) structures of chemical compounds. The CAR prediction model, which applies a 3D structure generator tool, called CORINA-generated and -optimized chemical structures, in the DeepSnap-DL demonstrated better performance than the existing methods using molecular descriptors. These results indicate that high performance in the prediction model using the DeepSnap-DL approach may be important to prepare suitable 3D chemical structures as input data and to enable the identification of modulators of the CAR.

In silico prediction of drug-induced developmental toxicity by using machine learning approaches

Some drugs and xenobiotics have the potential to disturb homeostasis, normal growth, differentiation, development or behavior during prenatal development or postnatally until puberty. Assessment of the developmental toxicity is one of the important safety considerations incorporated by international regulatory agencies. In this investigation, seven machine learning methods, including naïve Bayes, support vector machine, recursive partitioning, k-nearest neighbor, C4.5 decision tree, random forest and Adaboost, were used to build binary classification models for developmental toxicity. Among these models, the naïve Bayes classifier represented the best predictive performance and stability, which gave 91.11% overall prediction accuracy, 91.50% balanced accuracy and 0.818 MCC for the training set, and generated 83.93% concordance, 81.85% balanced accuracy and 0.627 MCC for the test set. The application domains were analyzed, and only one chemical in the test set was identified as outside the application domain. In addition, 10 important molecular descriptors related to developmental toxicity were selected by the genetic algorithm, which may contribute to explanation of the mechanisms of developmental toxicants. The best naïve Bayes classification model should be employed as alternative method for qualitative prediction of chemical-induced developmental toxicity in early stages of drug development.

Predicting estrogen receptor binding of chemicals using a suite of in silico methods - Complementary approaches of (Q)SAR, molecular docking and molecular dynamics

With the aim of obtaining reliable estimates of Estrogen Receptor (ER) binding for diverse classes of compounds, a weight of evidence approach using estimates from a suite of in silico models was assessed. The predictivity of a simple Majority Consensus of (Q)SAR models was assessed using a test set of compounds with experimental Relative Binding Affinity (RBA) data. Molecular docking was also carried out and the binding energies of these compounds to the ERα receptor were determined. For a few selected compounds, including a known full agonist and antagonist, the intrinsic activity was determined using low-mode molecular dynamics methods. Individual (Q)SAR model predictivity varied, as expected, with some models showing high sensitivity, others higher specificity. However, the Majority Consensus (Q)SAR prediction showed a high accuracy and reasonably balanced sensitivity and specificity. Molecular docking provided quantitative information on strength of binding to the ERα receptor. For the 50 highest binding affinity compounds with positive RBA experimental values, just 5 of them were predicted to be non-binders by the Majority QSAR Consensus. Furthermore, agonist-specific assay experimental values for these 5 compounds were negative, which indicates that they may be ER antagonists. We also showed different scenarios of combining (Q)SAR results with Molecular docking classification of ER binding based on cut-off values of binding energies, providing a rational combined strategy to maximize terms of toxicological interest.

An In Silico Approach for Assessment of the Membrane Transporter Activities of Phenols: A Case Study Based on Computational Models of Transport Activity for the Transporter Bilitranslocase

Phenols are the most abundant naturally accessible antioxidants present in a human normal diet. Since numerous beneficial applications of phenols as preventive agents in various diseases were revealed, the evaluation of phenols bioavailability is of high interest of researchers, consumers and drug manufacturers. The hydrophilic nature of phenols makes a cell membrane penetration difficult, which imply an alternative way of uptake via membrane transporters. However, the structural and functional data of membrane transporters are limited, thus the in silico modelling is really challenging and urgent tool in elucidation of transporter ligands. Focus of this research was a particular transporter bilitranslocase (BTL). BTL has a broad tissue expression (vascular endothelium, absorptive and excretory epithelia) and can transport wide variety of poly-aromatic compounds. With available BTL data (pKi [mmol/L] for 120 organic compounds) a robust and reliable QSAR models for BTL transport activity were developed and extrapolated on 300 phenolic compounds. For all compounds the transporter profiles were assessed and results show that dietary phenols and some drug candidates are likely to interact with BTL. Moreover, synopsis of predictions from BTL models and hits/predictions of 20 transporters from Metrabase and Chembench platforms were revealed. With such joint transporter analyses a new insights for elucidation of BTL functional role were acquired. Regarding limitation of models for virtual profiling of transporter interactions the computational approach reported in this study could be applied for further development of reliable in silico models for any transporter, if in vitro experimental data are available.

Classification models for identifying substances exhibiting acute contact toxicity in honeybees (Apis mellifera)$

Nowadays, environmental and biological endpoints can be predicted with in silico approaches if sufficient experimental data of good quality are available. Since the experimental evaluation of acute contact toxicity towards honeybees (Apis mellifera) is a complex and expensive assay, the computational models that follow OECD principles for this endpoint prediction represent important alternatives for safety prioritisation of chemicals, especially pesticides. We developed and validated counter-propagation artificial neural network (CPANN) models for in silico evaluation of toxicity of pesticides towards honeybees by using new in-house software. The data set included 254 pesticides with their toxicological experimental values (acute contact toxicity after 48 h of exposure - LD₅₀ [μg/bee]). The 2D structures of compounds were mathematically represented with 56 Dragon molecular descriptors (MDs). The two-category models were developed to separate compounds as toxic or non-toxic for two different thresholds: (i) toxic when LD₅₀ < 1 μg/bee and (ii) toxic when LD₅₀ < 100 μg/bee. The models give reliable predictions in an external validation set and cover a large structural space. They were applied to a structurally diverse data set of 395 experimentally untested pesticides; 19% of them were predicted as highly toxic towards bees.

The era of 3Rs implementation in developmental and reproductive toxicity (DART) testing: Current overview and future perspectives

Since adoption of the first globally implemented guidelines for developmental and reproductive toxicity (DART) testing for pharmaceuticals, industrial chemicals and agrochemicals, many years passed without major updates. However in recent years, significant changes in these guidelines have been made or are being implemented. These changes have been guided by the ethical drive to reduce, refine and replace (3R) animal testing, as well as the addition of endocrine disruptor relevant endpoints. Recent applied improvements have focused on reduction and refinement. Ongoing scientific and technical innovations will provide the means for replacement of animal testing in the future and will improve predictivity in humans. The aim of this review is to provide an overview of ongoing global DART endeavors in respect to the 3Rs, with an outlook towards future advances in DART testing aspiring to reduce animal testing to a minimum and the supreme ambition towards animal-free hazard and risk assessment.

The application of new HARD-descriptor available from the CORAL software to building up NOAEL models

Continuous QSAR models have been developed and validated for the prediction of no-observed-adverse-effect (NOAEL) in rats, using training and test sets from the Fraunhofer RepDose® database and EFSA's Chemical Hazards Database: OpenFoodTox. This paper demonstrates that the HARD index, as an integrated attribute of SMILES, improves the prediction power of NOAEL values using the continuous QSAR models and Monte Carlo simulations. The HARD-index is a line of eleven symbols, which represents the presence, or absence of eight chemical elements (nitrogen, oxygen, sulfur, phosphorus, fluorine, chlorine, bromine, and iodine) and different kinds of chemical bonds (double bond, triple bond, and stereo chemical bond). Optimal molecular descriptors calculated with the Monte Carlo technique (maximization of correlation coefficient between the descriptor and endpoint) give satisfactory predictive models for NOAEL. Optimal molecular descriptors calculated in this way with the Monte Carlo technique (maximization of correlation coefficient between the descriptor and endpoint) give amongst the best results available in the literature. The models are built up in accordance with OECD principles.