Improvement of quantitative structure-activity relationship (QSAR) tools for predicting Ames mutagenicity: outcomes of the Ames/QSAR International Challenge Project

AMPred-CNN: Ames mutagenicity prediction model based on convolutional neural networks

Mutagenicity assessment plays a pivotal role in the safety evaluation of chemicals, pharmaceuticals, and environmental compounds. In recent years, the development of robust computational models for predicting chemical mutagenicity has gained significant attention, driven by the need for efficient and cost-effective toxicity assessments. In this paper, we proposed AMPred-CNN, an innovative Ames mutagenicity prediction model based on Convolutional Neural Networks (CNNs), uniquely employing molecular structures as images to leverage CNNs' powerful feature extraction capabilities. The study employs the widely used benchmark mutagenicity dataset from Hansen et al. for model development and evaluation. Comparative analyses with traditional ML models on different molecular features reveal substantial performance enhancements. AMPred-CNN outshines these models, demonstrating superior accuracy, AUC, F1 score, MCC, sensitivity, and specificity on the test set. Notably, AMPred-CNN is further benchmarked against seven recent ML and DL models, consistently showcasing superior performance with an impressive AUC of 0.954. Our study highlights the effectiveness of CNNs in advancing mutagenicity prediction, paving the way for broader applications in toxicology and drug development.

Strategy proposal using QSAR models to approach mutagenicity assessment of non intentionally added substances in recycled plastic resins

CONTEXT

Transition to the use of recycled plastics raises an issue concerning safety assessment of Non Intentionally Added Substances (NIAS). To assess the mutagenic potential of the recycled polyethylene impurities and to evaluate the need to perform in vitro assays on recycled resins, this study lies in identifying existing NIAS associated with recycled Low/High Density Polyethylene and assessing the mutagenicity data-gaps by employing in silico tools.

METHODS

Quantitative Structure-Activity Relationship (QSAR) models predicting Ames mutagenicity were selected from literature, then NIAS were run to 1/evaluate performances of each model, 2/apply a QSAR strategy on the NIAS molecular space and address data-gaps.

RESULTS

Among the 165 NIAS identified, experimental Ames results were not found for 50 substances while the substances with experimental data were predominantly negatives. No individual model was able to predict all NIAS due to applicability domain limitations. Taking into account 1/calculated performances, 2/availability of applicability domain, 3/description of the Training Set, an Integrated Strategy was founded including Sarpy, Consensus and Protox to extend the applicability domain.

CONCLUSION & PERSPECTIVES

Existing data and predictions generated by this strategy suggest a low mutagenic potential of NIAS. Further investigation is needed to explore other genotoxicity mechanisms.

Progress in Predicting Ames Test Outcomes from Chemical Structures: An In-Depth Re-Evaluation of Models from the 1st and 2nd Ames/QSAR International Challenge Projects

The Ames/quantitative structure-activity relationship (QSAR) International Challenge Projects, held during 2014-2017 and 2020-2022, evaluated the performance of various predictive models. Despite the significant insights gained, the rules allowing participants to select prediction targets introduced ambiguity in model performance evaluation. This reanalysis identified the highest-performing prediction model, assuming a 100% coverage rate (COV) for all prediction target compounds and an estimated performance variation due to changes in COV. All models from both projects were evaluated using balance accuracy (BA), the Matthews correlation coefficient (MCC), the F1 score (F1), and the first principal component (PC1). After normalizing the COV, a correlation analysis with these indicators was conducted, and the evaluation index for all prediction models in terms of the COV was estimated. In total, using 109 models, the model with the highest estimated BA (76.9) at 100% COV was MMI-VOTE1, as reported by Meiji Pharmaceutical University (MPU). The best models for MCC, F1, and PC1 were all MMI-STK1, also reported by MPU. All the models reported by MPU ranked in the top four. MMI-STK1 was estimated to have F1 scores of 59.2, 61.5, and 63.1 at COV levels of 90%, 60%, and 30%, respectively. These findings highlight the current state and potential of the Ames prediction technology.

Mutagenicity assessment of two potential impurities in preparations of 5-amino-2,4,6 triiodoisophthalic acid, a key intermediate in the synthesis of the iodinated contrast agent iopamidol

5-Aminoisophthalic acid and 5-nitroisophthalic acid (5-NIPA) are potential impurities in preparations of 5-amino-2,4,6-triiodoisophthalic acid, which is a key intermediate in the synthesis of the iodinated contrast agent iopamidol. We have studied their mutagenicity in silico (quantitative structure-activity relationships, QSAR) and by the bacterial reverse mutation assay (Ames test). First, the compounds were screened with the tools Derek Nexus™ and Leadscope®. Both compounds were flagged as potentially mutagenic (class 3 under ICH M7). However, contrary to the in silico prediction, neither chemical was mutagenic in the Ames test (plate incorporation method) with or without S9 metabolic activation.

Development of a read-across-derived classification model for the predictions of mutagenicity data and its comparison with traditional QSAR models and expert systems

Mutagenicity is considered an important endpoint from the regulatory, environmental and medical points of view. Due to the wide number of compounds that may be of concern and the enormous expenses (in terms of time, money, and animals) associated with rodent mutagenicity bioassays, this endpoint is a major target for the development of alternative approaches for screening and prediction. The majority of old-aged expert systems and quantitative structure-activity relationship (QSAR) models may show reduced performance over time for their application on newer chemical candidates; thus, researchers constantly try to improve the modeling strategies. In our report, we initially performed traditional classification-based linear discriminant analysis (LDA) QSAR modeling using the benchmark Ames dataset of diverse chemicals (6512 compounds) to recognize the relationship between the molecules and their potential mutagenic behavior. The classical LDA QSAR model is developed from a selected set of 2D descriptors. The LDA QSAR model was developed by using a total of 31 descriptors identified from the analysis of the most discriminating features. Additionally, we have used similarity-derived features obtained from the read-across (RA) to develop an RA-based QSAR model. The developed RA-based LDA QSAR model has better predictivity, transferability, and interpretability compared to the LDA QSAR model, and it uses a very small number of descriptors compared to the classical QSAR model. Different machine learning (ML) models were also developed using the descriptors appearing in the read-across-based LDA QSAR model for comparative studies. We have checked the prediction quality of 216 true external set compounds using the novel similarity-derived RA model. The performance of the OECD toolbox is also compared with the RA-derived LDA QSAR model for a true external set. The current study aimed to explore the significance of the read-across-based algorithm and its application to the most current experimental mutagenicity data to complement already available expert systems.

Evaluation of QSAR models for predicting mutagenicity: outcome of the Second Ames/QSAR international challenge project

Quantitative structure-activity relationship (QSAR) models are powerful in silico tools for predicting the mutagenicity of unstable compounds, impurities and metabolites that are difficult to examine using the Ames test. Ideally, Ames/QSAR models for regulatory use should demonstrate high sensitivity, low false-negative rate and wide coverage of chemical space. To promote superior model development, the Division of Genetics and Mutagenesis, National Institute of Health Sciences, Japan (DGM/NIHS), conducted the Second Ames/QSAR International Challenge Project (2020-2022) as a successor to the First Project (2014-2017), with 21 teams from 11 countries participating. The DGM/NIHS provided a curated training dataset of approximately 12,000 chemicals and a trial dataset of approximately 1,600 chemicals, and each participating team predicted the Ames mutagenicity of each trial chemical using various Ames/QSAR models. The DGM/NIHS then provided the Ames test results for trial chemicals to assist in model improvement. Although overall model performance on the Second Project was not superior to that on the First, models from the eight teams participating in both projects achieved higher sensitivity than models from teams participating in only the Second Project. Thus, these evaluations have facilitated the development of QSAR models.

Assessing the utility of common arguments used in expert review of in silico predictions as part of ICH M7 assessments

Expert review of two predictions, made by complementary (quantitative) structure-activity relationship models, to an overall conclusion is a key component of using in silico tools to assess the mutagenic potential of impurities as part of the ICH M7 guideline. In lieu of a specified protocol, numerous publications have presented best practise guides, often indicating the occurrence of common prediction scenarios and the evidence required to resolve them. A semi-automated expert review tool has been implemented in Lhasa Limited's Nexus platform following collation of these common arguments and assignment to the associated prediction scenarios made by Derek Nexus and Sarah Nexus. Using datasets primarily donated by pharmaceutical companies, an automated analysis of the frequency these prediction scenarios occur, and the likelihood of the associated arguments assigning the correct resolution, could then be conducted. This article highlights that a relatively small number of common arguments may be used to accurately resolve many prediction scenarios to a single conclusion. The use of a standardised method of argumentation and assessment of evidence for a given impurity is proposed to improve the efficiency and consistency of expert review as part of an ICH M7 submission.

DeepAmes: A deep learning-powered Ames test predictive model with potential for regulatory application

The Ames assay is required by the regulatory agencies worldwide to assess the mutagenic potential risk of consumer products. As well as this in vitro assay, in silico approaches have been widely used to predict Ames test results as outlined in the International Council for Harmonization (ICH) guidelines. Building on this in silico approach, here we describe DeepAmes, a high performance and robust model developed with a novel deep learning (DL) approach for potential utility in regulatory science. DeepAmes was developed with a large and consistent Ames dataset (>10,000 compounds) and was compared with other five standard Machine Learning (ML) methods. Using a test set of 1,543 compounds, DeepAmes was the best performer in predicting the outcome of Ames assay. In addition, DeepAmes yielded the best and most stable performance up to when compounds were >30% outside of the applicability domain (AD). Regarding the potential for regulatory application, a revised version of DeepAmes with a much-improved sensitivity of 0.87 from 0.47. In conclusion, DeepAmes provides a DL-powered Ames test predictive model for predicting the results of Ames tests; with its defined AD and clear context of use, DeepAmes has potential for utility in regulatory application.

Datasets Construction and Development of QSAR Models for Predicting Micronucleus In Vitro and In Vivo Assay Outcomes

In silico (quantitative) structure-activity relationship modeling is an approach that provides a fast and cost-effective alternative to assess the genotoxic potential of chemicals. However, one of the limiting factors for model development is the availability of consolidated experimental datasets. In the present study, we collected experimental data on micronuclei in vitro and in vivo, utilizing databases and conducting a PubMed search, aided by text mining using the BioBERT large language model. Chemotype enrichment analysis on the updated datasets was performed to identify enriched substructures. Additionally, chemotypes common for both endpoints were found. Five machine learning models in combination with molecular descriptors, twelve fingerprints and two data balancing techniques were applied to construct individual models. The best-performing individual models were selected for the ensemble construction. The curated final dataset consists of 981 chemicals for micronuclei in vitro and 1309 for mouse micronuclei in vivo, respectively. Out of 18 chemotypes enriched in micronuclei in vitro, only 7 were found to be relevant for in vivo prediction. The ensemble model exhibited high accuracy and sensitivity when applied to an external test set of in vitro data. A good balanced predictive performance was also achieved for the micronucleus in vivo endpoint.

Multiple Instance Learning Improves Ames Mutagenicity Prediction for Problematic Molecular Species

The prediction of Ames mutagenicity continues to be a concern in both regulatory and pharmacological toxicology. Traditional quantitative structure-activity relationship (QSAR) models of mutagenicity make predictions based on molecular descriptors calculated on a chemical data set used in their training. However, it is known that molecules such as aromatic amines can be non-mutagenic themselves but metabolically activated by S9 rodent liver enzyme in Ames tests forming molecules such as iminoquinones or amine substituents that better stabilize mutagenic nitrenium ions in known pathways of mutagenicity. Modern in silico modeling methods can implicitly model these metabolites through consideration of the structural elements relevant to their formation but do not include explicit modeling of these metabolites' potential activity. These metabolites do not have a known individual mutagenicity label and, in their current state, cannot be fitted into a traditional QSAR model. Multiple instance learning (MIL) however can be applied to a group of metabolites and their parent under a single mutagenicity label. Here we trained MIL models on Ames data, first with an aromatic amines data set (n = 457), a class known to require metabolic activation, and subsequently on a larger data set (n = 6505) incorporating multiple molecular species. MIL was shown to be able to predict Ames mutagenicity with performance in line with previously established models (balanced accuracy = 0.778), suggesting its potential utility in Ames prediction applications. Furthermore, the MIL model predicted well on identified hard-to-predict molecule groups relative to the models in which these molecule groups were identified. These results are presumably due to the increased consideration of the metabolic contribution to the mutagenic outcome. Further exploration of MIL as a supplement to existing models could aid in the prediction of chemicals where implicit modeling of metabolites cannot fully grasp their characteristics. This paper demonstrates the potential of an MIL approach to modeling Ames tests with S9 and is particularly relevant to metabolically activated xenobiotic mutagens.

Exploring Dimensionality Reduction Techniques for Deep Learning Driven QSAR Models of Mutagenicity

Dimensionality reduction techniques are crucial for enabling deep learning driven quantitative structure-activity relationship (QSAR) models to navigate higher dimensional toxicological spaces, however the use of specific techniques is often arbitrary and poorly explored. Six dimensionality techniques (both linear and non-linear) were hence applied to a higher dimensionality mutagenicity dataset and compared in their ability to power a simple deep learning driven QSAR model, following grid searches for optimal hyperparameter values. It was found that comparatively simpler linear techniques, such as principal component analysis (PCA), were sufficient for enabling optimal QSAR model performances, which indicated that the original dataset was at least approximately linearly separable (in accordance with Cover's theorem). However certain non-linear techniques such as kernel PCA and autoencoders performed at closely comparable levels, while (especially in the case of autoencoders) being more widely applicable to potentially non-linearly separable datasets. Analysis of the chemical space, in terms of XLogP and molecular weight, uncovered that the vast majority of testing data occurred within the defined applicability domain, as well as that certain regions were measurably more problematic and antagonised performances. It was however indicated that certain dimensionality reduction techniques were able to facilitate uniquely beneficial navigations of the chemical space.

The VEGA Tool to Check the Applicability Domain Gives Greater Confidence in the Prediction of In Silico Models

A sound assessment of in silico models and their applicability domain can support the use of new approach methodologies (NAMs) in chemical risk assessment and requires increasing the users' confidence in this approach. Several approaches have been proposed to evaluate the applicability domain of such models, but their prediction power still needs a thorough assessment. In this context, the VEGA tool capable of assessing the applicability domain of in silico models is examined for a range of toxicological endpoints. The VEGA tool evaluates chemical structures and other features related to the predicted endpoints and is efficient in measuring applicability domain, enabling the user to identify less accurate predictions. This is demonstrated with many models addressing different endpoints, towards toxicity of relevance to human health, ecotoxicological endpoints, environmental fate, physicochemical and toxicokinetic properties, for both regression models and classifiers.

In-silico study for the identification of potential destabilizers between the spike protein of SARS-CoV-2 and human ACE-2

The emergence of the new SARS-CoV-2 virus, which causes the disease known as COVID-19, has generated a pandemic that has plunged the world into a health crisis. The infection process is triggered by the direct binding of the receptor-binding domain (RBD) of the spike (S) protein of SARS-CoV-2 to the angiotensin-converting enzyme 2 (ACE2) of the host cell. In the present study, virtual screening techniques such as molecular docking, molecular dynamics, calculation of free energy using the GBSA method, prediction of drug similarity, pharmacokinetic, and toxicological properties of various ligands interacting with the RBD-ACE2 complex were applied. The ligands radotinib, hinokiflavone, and ginkgetin were identified as potential destabilizers of the RBD-ACE2 interaction, which could produce their pharmacological effect by interacting at an allosteric site of ACE2, with affinity energy values of -10.2 ± 0.1, -9.8 ± 0.0, and -9.4 ± 0.0 kcal/mol, indicating strong receptor affinity. The complex with hinokiflavone showed the highest conformational stability and rigidity of the dynamic simulation and also obtained the best binding free energy of the three molecules, with an energy of -215.86 kcal/mol.

MicotoXilico: An Interactive Database to Predict Mutagenicity, Genotoxicity, and Carcinogenicity of Mycotoxins

Mycotoxins are secondary metabolites produced by certain filamentous fungi. They are common contaminants found in a wide variety of food matrices, thus representing a threat to public health, as they can be carcinogenic, mutagenic, or teratogenic, among other toxic effects. Several hundreds of mycotoxins have been reported, but only a few of them are regulated, due to the lack of data regarding their toxicity and mechanisms of action. Thus, a more comprehensive evaluation of the toxicity of mycotoxins found in foodstuffs is required. In silico toxicology approaches, such as Quantitative Structure-Activity Relationship (QSAR) models, can be used to rapidly assess chemical hazards by predicting different toxicological endpoints. In this work, for the first time, a comprehensive database containing 4360 mycotoxins classified in 170 categories was constructed. Then, specific robust QSAR models for the prediction of mutagenicity, genotoxicity, and carcinogenicity were generated, showing good accuracy, precision, sensitivity, and specificity. It must be highlighted that the developed QSAR models are compliant with the OECD regulatory criteria, and they can be used for regulatory purposes. Finally, all data were integrated into a web server that allows the exploration of the mycotoxin database and toxicity prediction. In conclusion, the developed tool is a valuable resource for scientists, industry, and regulatory agencies to screen the mutagenicity, genotoxicity, and carcinogenicity of non-regulated mycotoxins.

Augmenting Expert Knowledge-Based Toxicity Alerts by Statistically Mined Molecular Fragments

Structural alerts are molecular substructures assumed to be associated with molecular initiating events in various toxic effects and an integral part of in silico toxicology. However, alerts derived using the knowledge of human experts often suffer from a lack of predictivity, specificity, and satisfactory coverage. In this work, we present a method to build hybrid QSAR models by combining expert knowledge-based alerts and statistically mined molecular fragments. Our objective was to find out if the combination is better than the individual systems. Lasso regularization-based variable selection was applied on combined sets of knowledge-based alerts and molecular fragments, but the variable elimination was only allowed to happen on the molecular fragments. We tested the concept on three toxicity end points, i.e., skin sensitization, acute Daphnia toxicity, and Ames mutagenicity, which covered both classification and regression problems. Results showed the predictive performance of such hybrid models is, indeed, better than the models based solely on expert alerts or statistically mined fragments alone. The method also enables the discovery of activating and mitigating/deactivating features for toxicity alerts and the identification of new alerts, thereby reducing false positive and false negative outcomes commonly associated with generic alerts and alerts with poor coverage, respectively.

Prioritization of mycotoxins based on mutagenicity and carcinogenicity evaluation using combined in silico QSAR methods

Mycotoxins and their metabolites are a family of compounds that contains a great diversity of both structure and biological properties. Information on their toxicity is spread within several databases and in scientific literature. Due to the number of molecules and their structure diversity, the cost and time required for hazard evaluation of each compound is unrealistic. In that purpose, new approach methodologies (NAMs) can be applied to evaluate such large set of molecules. Among them, quantitative structure-activity relationship (QSAR) in silico models could be useful to predict the mutagenic and carcinogenic properties of mycotoxins. First, a complete list of 904 mycotoxins and metabolites was built. Then, some known mycotoxins were used to determine the best QSAR tools for mutagenicity and carcinogenicity predictions. The best tool was further applied to the whole list of 904 mycotoxins. At the end, 95 mycotoxins were identified as both mutagen and carcinogen and should be prioritized for further evaluation.

Artificial intelligence and real-world data for drug and food safety - A regulatory science perspective

In 2013, the Global Coalition for Regulatory Science Research (GCRSR) was established with members from over ten countries (www.gcrsr.net). One of the main objectives of GCRSR is to facilitate communication among global regulators on the rise of new technologies with regulatory applications through the annual conference Global Summit on Regulatory Science (GSRS). The 11th annual GSRS conference (GSRS21) focused on "Regulatory Sciences for Food/Drug Safety with Real-World Data (RWD) and Artificial Intelligence (AI)." The conference discussed current advancements in both AI and RWD approaches with a specific emphasis on how they impact regulatory sciences and how regulatory agencies across the globe are pursuing the adaptation and oversight of these technologies. There were presentations from Brazil, Canada, India, Italy, Japan, Germany, Switzerland, Singapore, the United Kingdom, and the United States. These presentations highlighted how various agencies are moving forward with these technologies by either improving the agencies' operation and/or preparing regulatory mechanisms to approve the products containing these innovations. To increase the content and discussion, the GSRS21 hosted two debate sessions on the question of "Is Regulatory Science Ready for AI?" and a workshop to showcase the analytical data tools that global regulatory agencies have been using and/or plan to apply to regulatory science. Several key topics were highlighted and discussed during the conference, such as the capabilities of AI and RWD to assist regulatory science policies for drug and food safety, the readiness of AI and data science to provide solutions for regulatory science. Discussions highlighted the need for a constant effort to evaluate emerging technologies for fit-for-purpose regulatory applications. The annual GSRS conferences offer a unique platform to facilitate discussion and collaboration across regulatory agencies, modernizing regulatory approaches, and harmonizing efforts.

Chemical rules for optimization of chemical mutagenicity via matched molecular pairs analysis and machine learning methods

Chemical mutagenicity is a serious issue that needs to be addressed in early drug discovery. Over a long period of time, medicinal chemists have manually summarized a series of empirical rules for the optimization of chemical mutagenicity. However, given the rising amount of data, it is getting more difficult for medicinal chemists to identify more comprehensive chemical rules behind the biochemical data. Herein, we integrated a large Ames mutagenicity data set with 8576 compounds to derive mutagenicity transformation rules for reversing Ames mutagenicity via matched molecular pairs analysis. A well-trained consensus model with a reasonable applicability domain was constructed, which showed favorable performance in the external validation set with an accuracy of 0.815. The model was used to assess the generalizability and validity of these mutagenicity transformation rules. The results demonstrated that these rules were of great value and could provide inspiration for the structural modifications of compounds with potential mutagenic effects. We also found that the local chemical environment of the attachment points of rules was critical for successful transformation. To facilitate the use of these mutagenicity transformation rules, we integrated them into ADMETopt2 ( http://lmmd.ecust.edu.cn/admetsar2/admetopt2/ ), a free web server for optimization of chemical ADMET properties. The above-mentioned approach would be extended to the optimization of other toxicity endpoints.

Machine Learning Toxicity Prediction: Latest Advances by Toxicity End Point

Machine learning (ML) models to predict the toxicity of small molecules have garnered great attention and have become widely used in recent years. Computational toxicity prediction is particularly advantageous in the early stages of drug discovery in order to filter out molecules with high probability of failing in clinical trials. This has been helped by the increase in the number of large toxicology databases available. However, being an area of recent application, a greater understanding of the scope and applicability of ML methods is still necessary. There are various kinds of toxic end points that have been predicted in silico. Acute oral toxicity, hepatotoxicity, cardiotoxicity, mutagenicity, and the 12 Tox21 data end points are among the most commonly investigated. Machine learning methods exhibit different performances on different data sets due to dissimilar complexity, class distributions, or chemical space covered, which makes it hard to compare the performance of algorithms over different toxic end points. The general pipeline to predict toxicity using ML has already been analyzed in various reviews. In this contribution, we focus on the recent progress in the area and the outstanding challenges, making a detailed description of the state-of-the-art models implemented for each toxic end point. The type of molecular representation, the algorithm, and the evaluation metric used in each research work are explained and analyzed. A detailed description of end points that are usually predicted, their clinical relevance, the available databases, and the challenges they bring to the field are also highlighted.

Multitask Deep Neural Networks for Ames Mutagenicity Prediction

The Ames mutagenicity test constitutes the most frequently used assay to estimate the mutagenic potential of drug candidates. While this test employs experimental results using various strains of Salmonella typhimurium, the vast majority of the published in silico models for predicting mutagenicity do not take into account the test results of the individual experiments conducted for each strain. Instead, such QSAR models are generally trained employing overall labels (i.e., mutagenic and nonmutagenic). Recently, neural-based models combined with multitask learning strategies have yielded interesting results in different domains, given their capabilities to model multitarget functions. In this scenario, we propose a novel neural-based QSAR model to predict mutagenicity that leverages experimental results from different strains involved in the Ames test by means of a multitask learning approach. To the best of our knowledge, the modeling strategy hereby proposed has not been applied to model Ames mutagenicity previously. The results yielded by our model surpass those obtained by single-task modeling strategies, such as models that predict the overall Ames label or ensemble models built from individual strains. For reproducibility and accessibility purposes, all source code and datasets used in our experiments are publicly available.

Toxicological assessment of SGLT2 inhibitors metabolites using in silico approach

Sodium-glucose cotransporter 2 inhibitors (SGLT2i) are the latest class of drugs approved to treat type 2 DM (T2DM). Although adverse effects are often caused by a metabolite rather than the drug itself, only the safety assessment of disproportionate drug metabolites is usually performed, which is of particular concern for drugs of chronic use, such as SGLT2i. Bearing this in mind, in silico tools are efficient strategies to reveal the risk assessment of metabolites, being endorsed by many regulatory agencies. Thereby, the goal of this study was to apply in silico methods to provide the metabolites toxicity assessment of the SGLT2i. Toxicological assessment from SGLT2i metabolites retrieved from the literature was estimated using the structure and/or statistical-based alert implemented in DataWarrior and ADMET predictorTM softwares. The drugs and their metabolites displayed no mutagenic, tumorigenic or cardiotoxic risks. Still, M1-2 and M3-1 were recognized as potential hepatotoxic compounds and M1-2, M1-3, M3-1, M3-2, M3-3 and M4-3, were estimated to have very toxic LD50 values in rats. All SGLT2i and the metabolites M3-4, M4-1 and M4-2, were predicted to have reproductive toxicity. These results support the awareness that metabolites may be potential mediators of drug-induced toxicities of the therapeutic agents.

Genotoxicity evaluation of nitrosamine impurities using human TK6 cells transduced with cytochrome P450s

Many nitrosamines are recognized as mutagens and potent rodent carcinogens. Over the past few years, nitrosamine impurities have been detected in various drugs leading to drug recalls. Although nitrosamines are included in a 'cohort of concern' because of their potential human health risks, most of this concern is based on rodent cancer and bacterial mutagenicity data, and there are little data on their genotoxicity in human-based systems. In this study, we employed human lymphoblastoid TK6 cells transduced with human cytochrome P450 (CYP) 2A6 to evaluate the genotoxicity of six nitrosamines that have been identified as impurities in drug products: N-nitrosodiethylamine (NDEA), N-nitrosoethylisopropylamine (NEIPA), N-nitroso-N-methyl-4-aminobutanoic acid (NMBA), N-nitrosomethylphenylamine (NMPA), N-nitrosodiisopropylamine (NDIPA), and N-nitrosodibutylamine (NDBA). Using flow cytometry-based assays, we found that 24-h treatment with NDEA, NEIPA, NMBA, and NMPA caused concentration-dependent increases in the phosphorylation of histone H2A.X (γH2A.X) in CYP2A6-expressing TK6 cells. Metabolism of these four nitrosamines by CYP2A6 also caused significant increases in micronucleus frequency as well as G2/M phase cell-cycle arrest. In addition, nuclear P53 activation was found in CYP2A6-expressing TK6 cells exposed to NDEA, NEIPA, and NMPA. Overall, the genotoxic potency of the six nitrosamine impurities in our test system was NMPA > NDEA ≈ NEIPA > NMBA > NDBA ≈ NDIPA. This study provides new information on the genotoxic potential of nitrosamines in human cells, complementing test results generated from traditional assays and partially addressing the issue of the relevance of nitrosamine genotoxicity for humans. The metabolically competent human cell system reported here may be a useful model for risk assessment of nitrosamine impurities found in drugs.

A local QSAR model based on the stability of nitrenium ions to support the ICH M7 expert review on the mutagenicity of primary aromatic amines

BACKGROUND

Aromatic amines, often used as intermediates for pharmaceutical synthesis, may be mutagenic and therefore pose a challenge as metabolites or impurities in drug development. However, predicting the mutagenicity of aromatic amines using commercially available, quantitative structure-activity relationship (QSAR) tools is difficult and often requires expert review. In this study, we developed a shareable QSAR tool based on nitrenium ion stability.

RESULTS

The evaluation using in-house aromatic amine intermediates revealed that our model has prediction accuracy of aromatic amine mutagenicity comparable to that of commercial QSAR tools. The effect of changing the number and position of substituents on the mutagenicity of aromatic amines was successfully explained by the change in the nitrenium ion stability. Furthermore, case studies showed that our QSAR tool can support the expert review with quantitative indicators.

CONCLUSIONS

This local QSAR tool will be useful as a quantitative support tool to explain the substituent effects on the mutagenicity of primary aromatic amines. By further refinement through method sharing and standardization, our tool can support the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) M7 expert review with quantitative indicators.

Using VEGAHUB Within a Weight-of-Evidence Strategy

Industrial needs and regulatory requirements have played a significant role in accelerating the use of nontesting methods including in silico tools as alternatives to animal testing. The main interest is not solely on the use of in silico tools, or in read-across, but on better toxicological safety assessment of substances, and for this purpose more advanced, integrated strategies have to be implemented. VEGAHUB wants to promote this broader view, not necessarily focused on a specific approach. Applying multiple tools and complementary approaches instead of one technique may provide more elements for a more robust evaluation, but at the same time it is important to have a conceptual scheme to integrate multiple, heterogeneous lines of evidence. We will show how the user can benefit from the diversity of tools available within the platform VEGAHUB for assessing the biological properties of chemical substances on an example of (non)mutagenicity.

In Silico Prediction of Chemically Induced Mutagenicity: A Weight of Evidence Approach Integrating Information from QSAR Models and Read-Across Predictions

Information on genotoxicity is an essential piece of information in the framework of several regulations aimed at evaluating chemical toxicity. In this context, QSAR models that can predict Ames genotoxicity can conveniently provide relevant information. Indeed, they can be straightforwardly and rapidly used for predicting the presence or absence of genotoxic hazards associated with the interactions of chemicals with DNA. Nevertheless, and despite their ease of use, the main interpretative challenge is related to a critical assessment of the information that can be gathered, thanks to these tools. This chapter provides guidance on how to use freely available QSAR and read-across tools provided by VEGA HUB and on how to interpret their predictions according to a weight-of-evidence approach.

MultiCASE Platform for In Silico Toxicology

Predictive and computational toxicology, a highly scientific and research-based field, is rapidly progressing with wider acceptance by regulatory agencies around the world. Almost every aspect of the field has seen fundamental changes during the last decade due to the availability of more data, usage, and acceptance of a variety of predictive tools and an increase in the overall awareness. Also, the influence from the recent explosive developments in the field of artificial intelligence has been significant. However, the need for sophisticated, easy to use and well-maintained software platforms for in silico toxicological assessments remains very high. The MultiCASE suite of software is one such platform that consists of an integrated collection of software programs, tools, and databases. While providing easy-to-use and highly useful tools that are relevant at present, it has always remained at the forefront of research and development by inventing new technologies and discovering new insights in the area of QSAR, artificial intelligence, and machine learning. This chapter gives the background, an overview of the software and databases involved, and a brief description of the usage methodology with the aid of examples.

Artificial Intelligence, Machine Learning, and Deep Learning in Real-Life Drug Design Cases

The discovery and development of drugs is a long and expensive process with a high attrition rate. Computational drug discovery contributes to ligand discovery and optimization, by using models that describe the properties of ligands and their interactions with biological targets. In recent years, artificial intelligence (AI) has made remarkable modeling progress, driven by new algorithms and by the increase in computing power and storage capacities, which allow the processing of large amounts of data in a short time. This review provides the current state of the art of AI methods applied to drug discovery, with a focus on structure- and ligand-based virtual screening, library design and high-throughput analysis, drug repurposing and drug sensitivity, de novo design, chemical reactions and synthetic accessibility, ADMET, and quantum mechanics.

A Classification Model to Identify Direct-Acting Mutagenic Polycyclic Aromatic Hydrocarbon Transformation Products

Polycyclic aromatic hydrocarbons (PAHs) are a complex group of environmental contaminants, many having long environmental half-lives. As these compounds degrade, the changes in their structure can result in a substantial increase in mutagenicity compared to the parent compound. Over time, each individual PAH can potentially degrade into several thousand unique transformation products, creating a complex, constantly evolving set of intermediates. Microbial degradation is the primary mechanism of their transformation and ultimate removal from the environment, and this process can result in mutagenic activation similar to the metabolic activation that can occur in multicellular organisms. The diversity of the potential intermediate structures in PAH-contaminated environments renders hazard assessment difficult for both remediation professionals and regulators. A mixture of structural and energetic descriptors has proven effective in existing studies for classifying which PAH transformation products will be mutagenic. However, most existing studies of environmental PAH mutagens primarily focus on nitrogenated derivatives, which are prevalent in the atmosphere and not as relevant in soil. Additionally, PAH products commonly found in the environment can range from as large as five rings to as small as a single ring, requiring a broadly inclusive methodology to comprehensively evaluate mutagenic potential. We developed a combination of supervised and unsupervised machine learning methods to predict environmentally induced PAH mutagenicity with improved performance over currently available tools. K-means clustering with principal component analysis allows us to identify molecular clusters that we hypothesize to have similar mechanisms of action. Recursive feature elimination identifies the most influential descriptors. The cluster-specific regression outperforms available classifiers in predicting direct-acting mutagens resulting from the microbial biodegradation of PAHs and provides direction for future studies evaluating the environmental hazards resulting from PAH biodegradation.

Machine learning - Predicting Ames mutagenicity of small molecules

In modern drug discovery, detection of a compound's potential mutagenicity is crucial. However, the traditional method of mutagenicity detection using the Ames test is costly and time consuming as the compounds need to be synthesised and then tested and the results are not always accurate and reproducible. Therefore, it would be advantageous to develop robust in silico models which can accurately predict the mutagenicity of a compound prior to synthesis to overcome the inadequacies of the Ames test. After curation of a previously defined compound mutagenicity library, over 5000 molecules had their chemical fingerprints and molecular properties calculated. Using 8 classification modelling algorithms, including support vector machine (SVM), random forest (RF) and extreme gradient boosting (XGB), a total of 112 predictive models have been constructed. Their performance has been assessed using 10-fold cross validation and a hold-out test set and some of the top performing models have been assessed using the y-randomisation approach. As a result, we have found SVM and XGB models to have good performance during the 10-fold cross validation (AUROC >0.90, sensitivity >0.85, specificity >0.75, balanced accuracy >0.80, Kappa >0.65) and on the test set (AUROC >0.65, sensitivity >0.65, specificity >0.60, balanced accuracy >0.65, Kappa >0.30). We have also identified molecular properties that are the most influential for mutagenicity prediction when combined with chemical molecular fingerprints. Using the Class A mutagenic compounds from the Ames/QSAR International Challenge Project, we were able to verify our models perform better, predicting more mutagens correctly then the StarDrop Ames mutagenicity prediction and TEST mutagenicity prediction.

In silico assessment of genotoxicity. Combinations of sensitive structural alerts minimize false negative predictions for all genotoxicity endpoints and can single out chemicals for which experimentation can be avoided

Genotoxicity assessment of chemicals has a crucial role in most regulations. Due to labor, time, cost, and animal welfare issues, attention is being given to (Q)SAR methods. A strategic application of alternative methods is to first use a sequence of conservative (very sensitive) (Q)SARs and/or in vitro models to arrive at the conclusion that no further testing is necessary for negatives, and to use mechanistically based, Weight-Of-Evidence approach to evaluate the chemicals showing positive results. The ICH M7 guideline to detect DNA-reactive impurities in drugs follows these lines (recommending solely (Q)SAR in step 1). However, ICH M7 focuses only on Ames test. Here a large database of more than 6000 chemicals positive in at least one endpoint (in vitro gene mutations or chromosomal aberrations, in vivo micronucleus, aneugenicity) were analyzed with structural alerts implemented in the OECD QSAR Toolbox, resulting in maximum 3% false negatives. These promising results indicate that it may be possible to extend the approach to the whole range of genotoxicity endpoints required by regulations. Since structural alerts may generate false positives, cautious follow-up of positives is recommended (with e.g., statistically based QSARs, read across of similar chemicals, expert judgement, and experimentation when necessary).

An overview of current practices for regulatory risk assessment with lessons learnt from cosmetics in the European Union

Risk assessments of various types of chemical compounds are carried out in the European Union (EU) foremost to comply with legislation and to support regulatory decision-making with respect to their safety. Historically, risk assessment has relied heavily on animal experiments. However, the EU is committed to reduce animal experimentation and has implemented several legislative changes, which have triggered a paradigm shift towards human-relevant animal-free testing in the field of toxicology, in particular for risk assessment. For some specific endpoints, such as skin corrosion and irritation, validated alternatives are available whilst for other endpoints, including repeated dose systemic toxicity, the use of animal data is still central to meet the information requirements stipulated in the different legislations. The present review aims to provide an overview of established and more recently introduced methods for hazard assessment and risk characterisation for human health, in particular in the context of the EU Cosmetics Regulation (EC No 1223/2009) as well as the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) Regulation (EC 1907/2006).

AI in drug development: a multidisciplinary perspective

The introduction of a new drug to the commercial market follows a complex and long process that typically spans over several years and entails large monetary costs due to a high attrition rate. Because of this, there is an urgent need to improve this process using innovative technologies such as artificial intelligence (AI). Different AI tools are being applied to support all four steps of the drug development process (basic research for drug discovery; pre-clinical phase; clinical phase; and postmarketing). Some of the main tasks where AI has proven useful include identifying molecular targets, searching for hit and lead compounds, synthesising drug-like compounds and predicting ADME-Tox. This review, on the one hand, brings in a mathematical vision of some of the key AI methods used in drug development closer to medicinal chemists and, on the other hand, brings the drug development process and the use of different models closer to mathematicians. Emphasis is placed on two aspects not mentioned in similar surveys, namely, Bayesian approaches and their applications to molecular modelling and the eventual final use of the methods to actually support decisions. Promoting a perfect synergy.

Toward regulatory acceptance and improving the prediction confidence of in silico approaches: a case study of genotoxicity

Introduction: Genotoxicity is an imperative component of the human health safety assessment of chemicals. Its secure forecast is of the utmost importance for all health prevention strategies and regulations.Areas covered: We surveyed several types of alternative, animal-free approaches ((quantitative) structure-activity relationship (Q)SAR, read-across, Adverse Outcome Pathway, Integrated Approaches to Testing and Assessment) for genotoxicity prediction within the needs of regulatory frameworks, putting special emphasis on data quality and uncertainties issues.Expert opinion: (Q)SAR models and read-across approaches for in vitro bacterial mutagenicity have sufficient reliability for use in prioritization processes, and as support in regulatory decisions in combination with other types of evidence. (Q)SARs and read-across methodologies for other genotoxicity endpoints need further improvements and should be applied with caution. It appears that there is still large room for improvement of genotoxicity prediction methods. Availability of well-curated high-quality databases, covering a broader chemical space, is one of the most important needs. Integration of in silico predictions with expert knowledge, weight-of-evidence-based assessment, and mechanistic understanding of genotoxicity pathways are other key points to be addressed for the generation of more accurate and trustable results.

QSAR modeling without descriptors using graph convolutional neural networks: the case of mutagenicity prediction

Deep neural networks are effective in learning directly from low-level encoded data without the need of feature extraction. This paper shows how QSAR models can be constructed from 2D molecular graphs without computing chemical descriptors. Two graph convolutional neural network-based models are presented with and without a Bayesian estimation of the prediction uncertainty. The property under investigation is mutagenicity: Models developed here predict the output of the Ames test. These models take the SMILES representation of the molecules as input to produce molecular graphs in terms of adjacency matrices and subsequently use attention mechanisms to weight the role of their subgraphs in producing the output. The results positively compare with current state-of-the-art models. Furthermore, our proposed model interpretation can be enhanced by the automatic extraction of the substructures most important in driving the prediction, as well as by uncertainty estimations.

Development of a new quantitative structure-activity relationship model for predicting Ames mutagenicity of food flavor chemicals using StarDrop™ auto-Modeller™

BACKGROUND

Food flavors are relatively low molecular weight chemicals with unique odor-related functional groups that may also be associated with mutagenicity. These chemicals are often difficult to test for mutagenicity by the Ames test because of their low production and peculiar odor. Therefore, application of the quantitative structure-activity relationship (QSAR) approach is being considered. We used the StarDrop™ Auto-Modeller™ to develop a new QSAR model.

RESULTS

In the first step, we developed a new robust Ames database of 406 food flavor chemicals consisting of existing Ames flavor chemical data and newly acquired Ames test data. Ames results for some existing flavor chemicals have been revised by expert reviews. We also collected 428 Ames test datasets for industrial chemicals from other databases that are structurally similar to flavor chemicals. A total of 834 chemicals' Ames test datasets were used to develop the new QSAR models. We repeated the development and verification of prototypes by selecting appropriate modeling methods and descriptors and developed a local QSAR model. A new QSAR model "StarDrop NIHS 834_67" showed excellent performance (sensitivity: 79.5%, specificity: 96.4%, accuracy: 94.6%) for predicting Ames mutagenicity of 406 food flavors and was better than other commercial QSAR tools.

CONCLUSIONS

A local QSAR model, StarDrop NIHS 834_67, was customized to predict the Ames mutagenicity of food flavor chemicals and other low molecular weight chemicals. The model can be used to assess the mutagenicity of food flavors without actual testing.

Quantitative structure-activity relationship models for genotoxicity prediction based on combination evaluation strategies for toxicological alternative experiments

Mutagenicity exerts adverse effects on humans. Conventional methods cannot simultaneously predict the toxicity of a large number of compounds. Most mutagenicity prediction models are based on a single experimental type and lack other experimental combination data as support, resulting in limited application scope and predictive ability. In this study, we partitioned data from GENE-TOX, CPDB, and Chemical Carcinogenesis Research Information System according to the weight-of-evidence method for modelling. In our data set, in vivo and in vitro experiments in groups as well as prokaryotic and eukaryotic cell experiments were included in accordance with the ICH guideline. We compared the two experimental combinations mentioned in the weight-of-evidence method and reintegrated the experimental data into three groups. Nine sub-models and three fusion models were established using random forest (RF), support vector machine (SVM), and back propagation (BP) neural network algorithms. When fusing base models under the same algorithm according to the ensemble rules, all models showed excellent predictive performance. The RF, SVM, and BP fusion models reached a prediction accuracy rate of 83.4%, 80.5%, 79.0% respectively. The area under the curve (AUC) reached 0.853, 0.897, 0.865 respectively. Therefore, the established fusion QSAR models can serve as an early warning system for mutagenicity of compounds.

Multi-Time-Scale Features for Accurate Respiratory Sound Classification

The COVID-19 pandemic has amplified the urgency of the developments in computer-assisted medicine and, in particular, the need for automated tools supporting the clinical diagnosis and assessment of respiratory symptoms. This need was already clear to the scientific community, which launched an international challenge in 2017 at the International Conference on Biomedical Health Informatics (ICBHI) for the implementation of accurate algorithms for the classification of respiratory sound. In this work, we present a framework for respiratory sound classification based on two different kinds of features: (i) short-term features which summarize sound properties on a time scale of tenths of a second and (ii) long-term features which assess sounds properties on a time scale of seconds. Using the publicly available dataset provided by ICBHI, we cross-validated the classification performance of a neural network model over 6895 respiratory cycles and 126 subjects. The proposed model reached an accuracy of 85%±3% and an precision of 80%±8%, which compare well with the body of literature. The robustness of the predictions was assessed by comparison with state-of-the-art machine learning tools, such as the support vector machine, Random Forest and deep neural networks. The model presented here is therefore suitable for large-scale applications and for adoption in clinical practice. Finally, an interesting observation is that both short-term and long-term features are necessary for accurate classification, which could be the subject of future studies related to its clinical interpretation.

Screening for Ames mutagenicity of food flavor chemicals by (quantitative) structure-activity relationship

Background

(Quantitative) Structure-Activity Relationship ((Q)SAR) is a promising approach to predict the potential adverse effects of chemicals based on their structure without performing toxicological studies. We evaluate the mutagenicity of food flavor chemicals by (Q) SAR tools, identify potentially mutagenic chemicals, and verify their mutagenicity by actual Ames test.

Results

The Ames mutagenicity of 3942 food flavor chemicals was predicted using two (Q)SAR) tools, DEREK Nexus and CASE Ultra. Three thousand five hundred seventy-five chemicals (91%) were judged to be negative in both (Q) SAR tools, and 75 chemicals (2%) were predicted to be positive in both (Q) SAR tools. When the Ames test was conducted on ten of these positive chemicals, nine showed positive results.

Conclusion

The (Q) SAR method can be used for screening the mutagenicity of food flavors.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41021-020-00171-1.

Identification of flavouring substances of genotoxic concern present in e-cigarette refills

E-cigarettes have become very popular, a trend that has been stimulated by the wide variety of available e-liquid flavours. Considering the large number of e-liquid flavours (>7000), there is an urgent need to establish a screening strategy to prioritize the flavouring substances of highest concern for human health. In the present study, a prioritization strategy combining analytical screening, in silico tools and literature data was developed to identify potentially genotoxic e-liquid flavourings. Based on the analysis of 129 e-liquids collected on the Belgian market, 60 flavourings with positive in silico predictions for genotoxicity were identified. By using literature data, genotoxicity was excluded for 33 of them whereas for 5, i.e. estragole, safrole, 2-furylmethylketon, 2,5-dimethyl-4-hydroxyl-3(2H)-furanone and transhexanal, there was a clear concern for in vivo genotoxicity. A selection of 4 out of the remaining 22 flavourings was tested in two in vitro genotoxicity assays. Three out of the four tested flavourings induced gene mutations and chromosome damage in vitro, whereas equivocal results were obtained for the fourth compound. Thus, although there is a legislative framework which excludes the use of CMR compounds in e-liquids, flavourings of genotoxic concern are present and might pose a health risk for e-cigarette users.

(Q)SAR tools for the prediction of mutagenic properties: Are they ready for application in pesticide regulation?

The assessment of human health risks resulting from the presence of metabolites in groundwater and food residues has become an important element in pesticide authorisation. In this context, the evaluation of mutagenicity is of particular interest and a paradigm shift from exposure-triggered testing to in silico-based screening has been recommended in the European Food Safety Authority (EFSA) Guidance on the establishment of the residue definition for dietary risk assessment. In addition, it is proposed to apply in silico predictions when experimental mutagenicity testing is not possible due to a lack of sufficient quantities of the pesticide metabolite. This, combined with animal welfare and economic considerations, has led to a situation where an increasing number of in silico studies are submitted to regulatory authorities. Whilst there is extensive experience with in silico predictions for mutagenicity in the chemical and pharmaceutical industry, their suitability in pesticide regulation is still insufficiently considered. Therefore, we herein discuss critical issues that need to be resolved to successfully implement (Quantitative) Structure-Activity Relationship ((Q)SAR) as an accepted tool in pesticide regulation. For illustration purposes, the results of a pilot study are included. The presented study highlights a need for further improvement regarding the predictivity and applicability domain of (Q)SAR systems for pesticides and their metabolites, but also raises other questions such as model selection, establishment of acceptance criteria, harmonised approaches to the combination of model outputs into overall conclusions, adequate reporting and data sharing. © 2020 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Density Functional Theory in the Prediction of Mutagenicity: A Perspective

As a field, computational toxicology is concerned with using in silico models to predict and understand the origins of toxicity. It is fast, relatively inexpensive, and avoids the ethical conundrum of using animals in scientific experimentation. In this perspective, we discuss the importance of computational models in toxicology, with a specific focus on the different model types that can be used in predictive toxicological approaches toward mutagenicity (SARs and QSARs). We then focus on how quantum chemical methods, such as density functional theory (DFT), have previously been used in the prediction of mutagenicity. It is then discussed how DFT allows for the development of new chemical descriptors that focus on capturing the steric and energetic effects that influence toxicological reactions. We hope to demonstrate the role that DFT plays in understanding the fundamental, intrinsic chemistry of toxicological reactions in predictive toxicology.