Machine Learning Methods in Computational Toxicology

An in silico investigation of the toxicological effects and biological activities of 3-phenoxybenzoic acid and its metabolite products

We aimed to elucidate the toxic effects and biological activities of 3-phenoxybenzoic acid (3PBA) and its metabolite products. Numerous in silico methods were used to identify the toxic effects and biological activities of 3PBA, including PASS online, molecular docking, ADMETlab 2.0, ADMESWISS, MetaTox, and molecular dynamic simulation. Ten metabolite products were identified via Phase II reactions (O-glucuronidation, O-sulfation, and methylation). All of the investigated compounds were followed by Lipinski's rule, indicating that they were stimulants or inducers of hazardous processes. Because of their high gastrointestinal absorption and ability to reach the blood-brain barrier, the studied compounds' physicochemical and pharmacokinetic properties matched existing evidence of harmful effects, including hematemesis, reproductive dysfunction, allergic dermatitis, toxic respiration, and neurotoxicity. The studied compounds have been linked to the apoptotic pathway, the reproductivity system, neuroendocrine disruptors, phospholipid-translocating ATPase inhibitors, and JAK2 expression. An O-glucuronidation metabolite product demonstrated higher binding affinity and interaction with CYP2C9, CYP3A4, caspase 3, and caspase 8 than 3PBA and other metabolite products, whereas metabolite products from methylation were predominant and more toxic. Our in silico findings partly meet the 3Rs principle by minimizing animal testing before more study is needed to identify the detrimental effects of 3PBA on other organs (liver, kidneys). Future research directions may involve experimental validation of in silico predictions, elucidation of molecular mechanisms, and exploration of therapeutic interventions. These findings contribute to our understanding of the toxicological profile of 3PBA and its metabolites, which has implications for risk assessment and regulatory decisions.

Optimized Bags of Artificial Neural Networks Can Predict the Viability of Organisms Exposed to Nanoparticles

Prediction of organismal viability upon exposure to a nanoparticle in varying environments─as fully specified at the molecular scale─has emerged as a useful figure of merit in the design of engineered nanoparticles. We build on our earlier finding that a bag of artificial neural networks (ANNs) can provide such a prediction when such machines are trained with a relatively small data set (with ca. 200 examples). Therein, viabilities were predicted by consensus using the weighted means of the predictions from the bags. Here, we confirm the accuracy and precision of the prediction of nanoparticle viabilities using an optimized bag of ANNs over sets of data examples that had not previously been used in the training and validation process. We also introduce the viability strip, rather than a single value, as the prediction and construct it from the viability probability distribution of an ensemble of ANNs compatible with the data set. Specifically, the ensemble consists of the ANNs arising from subsets of the data set corresponding to different splittings between training and validation, and the different bags (k-folds). A k - 1 k machine uses a single partition (or bag) of k - 1 ANNs each trained on 1/k of the data to obtain a consensus prediction, and a k-bag machine quorum samples the k possible k - 1 k machines available for a given partition. We find that with increasing k in the k-bag or k - 1 k machines, the viability strips become more normally distributed and their predictions become more precise. Benchmark comparisons between ensembles of 4-bag machines and 3 4 fraction machines suggest that the 3 4 fraction machine has similar accuracy while overcoming some of the challenges arising from divergent ANNs in the 4-bag machines.

Artificial intelligence (AI)-it's the end of the tox as we know it (and I feel fine)

The rapid progress of AI impacts diverse scientific disciplines, including toxicology, and has the potential to transform chemical safety evaluation. Toxicology has evolved from an empirical science focused on observing apical outcomes of chemical exposure, to a data-rich field ripe for AI integration. The volume, variety and velocity of toxicological data from legacy studies, literature, high-throughput assays, sensor technologies and omics approaches create opportunities but also complexities that AI can help address. In particular, machine learning is well suited to handle and integrate large, heterogeneous datasets that are both structured and unstructured-a key challenge in modern toxicology. AI methods like deep neural networks, large language models, and natural language processing have successfully predicted toxicity endpoints, analyzed high-throughput data, extracted facts from literature, and generated synthetic data. Beyond automating data capture, analysis, and prediction, AI techniques show promise for accelerating quantitative risk assessment by providing probabilistic outputs to capture uncertainties. AI also enables explanation methods to unravel mechanisms and increase trust in modeled predictions. However, issues like model interpretability, data biases, and transparency currently limit regulatory endorsement of AI. Multidisciplinary collaboration is needed to ensure development of interpretable, robust, and human-centered AI systems. Rather than just automating human tasks at scale, transformative AI can catalyze innovation in how evidence is gathered, data are generated, hypotheses are formed and tested, and tasks are performed to usher new paradigms in chemical safety assessment. Used judiciously, AI has immense potential to advance toxicology into a more predictive, mechanism-based, and evidence-integrated scientific discipline to better safeguard human and environmental wellbeing across diverse populations.

Pharmacological Properties of Chemically Characterized Extracts from Mastic Tree: In Vitro and In Silico Assays

The mastic tree, scientifically known as Pistacia lentiscus, which belongs to the Anacardiaceae family, was used in this study. The aim of this research was to analyze the chemical composition of this plant and assess its antioxidant and antibacterial properties using both laboratory experiments and computer simulations through molecular docking, a method that predicts the binding strength of a small molecule to a protein. The soxhlet method (SE) was employed to extract substances from the leaves of P. lentiscus found in the eastern region of Morocco. Hexane and methanol were the solvents used for the extraction process. The n-hexane extract was subjected to gas chromatography-mass spectrometry (GC/MS) to identify its fatty acid content. The methanolic extract underwent high-performance liquid chromatography with a diode-array detector (HPLC-DAD) to determine the presence of phenolic compounds. Antioxidant activity was assessed using the DPPH spectrophotometric test. The findings revealed that the main components in the n-hexane extract were linoleic acid (40.97 ± 0.33%), oleic acid (23.69 ± 0.12%), and palmitic acid (22.83 ± 0.10%). Catechin (37.05 ± 0.15%) was identified as the predominant compound in the methanolic extract through HPLC analysis. The methanolic extract exhibited significant DPPH radical scavenging, with an IC50 value of 0.26 ± 0.14 mg/mL. The antibacterial activity was tested against Staphylococcus aureus, Listeria innocua, and Escherichia coli, while the antifungal activity was evaluated against Geotrichum candidum and Rhodotorula glutinis. The P. lentiscus extract demonstrated notable antimicrobial effects. Additionally, apart from molecular docking, other important factors, such as drug similarity, drug metabolism and distribution within the body, potential adverse effects, and impact on bodily systems, were considered for the substances derived from P. lentiscus. Scientific algorithms, such as Prediction of Activity Spectra for Substances (PASS), Absorption, Distribution, Metabolism, Excretion (ADME), and Pro-Tox II, were utilized for this assessment. The results obtained from this research support the traditional medicinal usage of P. lentiscus and suggest its potential for drug development.

Artificial intelligence to estimate the tear film breakup time and diagnose dry eye disease

The use of artificial intelligence (AI) in the diagnosis of dry eye disease (DED) remains limited due to the lack of standardized image formats and analysis models. To overcome these issues, we used the Smart Eye Camera (SEC), a video-recordable slit-lamp device, and collected videos of the anterior segment of the eye. This study aimed to evaluate the accuracy of the AI algorithm in estimating the tear film breakup time and apply this model for the diagnosis of DED according to the Asia Dry Eye Society (ADES) DED diagnostic criteria. Using the retrospectively corrected DED videos of 158 eyes from 79 patients, 22,172 frames were annotated by the DED specialist to label whether or not the frame had breakup. The AI algorithm was developed using the training dataset and machine learning. The DED criteria of the ADES was used to determine the diagnostic performance. The accuracy of tear film breakup time estimation was 0.789 (95% confidence interval (CI) 0.769-0.809), and the area under the receiver operating characteristic curve of this AI model was 0.877 (95% CI 0.861-0.893). The sensitivity and specificity of this AI model for the diagnosis of DED was 0.778 (95% CI 0.572-0.912) and 0.857 (95% CI 0.564-0.866), respectively. We successfully developed a novel AI-based diagnostic model for DED. Our diagnostic model has the potential to enable ophthalmology examination outside hospitals and clinics.

Automated detection of toxicophores and prediction of mutagenicity using PMCSFG algorithm

Maximum common substructures (MCS) have received a lot of attention in the chemoinformatics community. They are typically used as a similarity measure between molecules, showing high predictive performance when used in classification tasks, while being easily explainable substructures. In the present work, we applied the Pairwise Maximum Common Subgraph Feature Generation (PMCSFG) algorithm to automatically detect toxicophores (structural alerts) and to compute fingerprints based on MCS. We present a comparison between our MCS-based fingerprints and 12 well-known chemical fingerprints when used as features in machine learning models. We provide an experimental evaluation and discuss the usefulness of the different methods on mutagenicity data. The features generated by the MCS method have a state-of-the-art performance when predicting mutagenicity, while they are more interpretable than the traditional chemical fingerprints.

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.

Machine learning models for rat multigeneration reproductive toxicity prediction

Reproductive toxicity is one of the prominent endpoints in the risk assessment of environmental and industrial chemicals. Due to the complexity of the reproductive system, traditional reproductive toxicity testing in animals, especially guideline multigeneration reproductive toxicity studies, take a long time and are expensive. Therefore, machine learning, as a promising alternative approach, should be considered when evaluating the reproductive toxicity of chemicals. We curated rat multigeneration reproductive toxicity testing data of 275 chemicals from ToxRefDB (Toxicity Reference Database) and developed predictive models using seven machine learning algorithms (decision tree, decision forest, random forest, k-nearest neighbors, support vector machine, linear discriminant analysis, and logistic regression). A consensus model was built based on the seven individual models. An external validation set was curated from the COSMOS database and the literature. The performances of individual and consensus models were evaluated using 500 iterations of 5-fold cross-validations and the external validation data set. The balanced accuracy of the models ranged from 58% to 65% in the 5-fold cross-validations and 45%–61% in the external validations. Prediction confidence analysis was conducted to provide additional information for more appropriate applications of the developed models. The impact of our findings is in increasing confidence in machine learning models. We demonstrate the importance of using consensus models for harnessing the benefits of multiple machine learning models (i.e., using redundant systems to check validity of outcomes). While we continue to build upon the models to better characterize weak toxicants, there is current utility in saving resources by being able to screen out strong reproductive toxicants before investing in vivo testing. The modeling approach (machine learning models) is offered for assessing the rat multigeneration reproductive toxicity of chemicals. Our results suggest that machine learning may be a promising alternative approach to evaluate the potential reproductive toxicity of chemicals.

Machine learning and artificial intelligence in toxicological sciences

Machine learning and artificial intelligence approaches have revolutionized multiple disciplines, including toxicology. This review summarizes representative recent applications of machine learning and artificial intelligence approaches in different areas of toxicology, including physiologically based pharmacokinetic (PBPK) modeling, quantitative structure-activity relationship modeling for toxicity prediction, adverse outcome pathway analysis, high-throughput screening, toxicogenomics, big data and toxicological databases. By leveraging machine learning and artificial intelligence approaches, now it is possible to develop PBPK models for hundreds of chemicals efficiently, to create in silico models to predict toxicity for a large number of chemicals with similar accuracies compared to in vivo animal experiments, and to analyze a large amount of different types of data (toxicogenomics, high-content image data, etc.) to generate new insights into toxicity mechanisms rapidly, which was impossible by manual approaches in the past. To continue advancing the field of toxicological sciences, several challenges should be considered: (1) not all machine learning models are equally useful for a particular type of toxicology data, and thus it is important to test different methods to determine the optimal approach; (2) current toxicity prediction is mainly on bioactivity classification (yes/no), so additional studies are needed to predict the intensity of effect or dose-response relationship; (3) as more data become available, it is crucial to perform rigorous data quality check and develop infrastructure to store, share, analyze, evaluate, and manage big data; and (4) it is important to convert machine learning models to user-friendly interfaces to facilitate their applications by both computational and bench scientists.

Predicting Nanoparticle Delivery to Tumors Using Machine Learning and Artificial Intelligence Approaches

Background

Low delivery efficiency of nanoparticles (NPs) to the tumor is a critical barrier in the field of cancer nanomedicine. Strategies on how to improve NP tumor delivery efficiency remain to be determined.

Methods

This study analyzed the roles of NP physicochemical properties, tumor models, and cancer types in NP tumor delivery efficiency using multiple machine learning and artificial intelligence methods, using data from a recently published Nano-Tumor Database that contains 376 datasets generated from a physiologically based pharmacokinetic (PBPK) model.

Results

The deep neural network model adequately predicted the delivery efficiency of different NPs to different tumors and it outperformed all other machine learning methods; including random forest, support vector machine, linear regression, and bagged model methods. The adjusted determination coefficients (R2) in the full training dataset were 0.92, 0.77, 0.77 and 0.76 for the maximum delivery efficiency (DEmax), delivery efficiency at 24 h (DE24), at 168 h (DE168), and at the last sampling time (DETlast). The corresponding R2 values in the test dataset were 0.70, 0.46, 0.33 and 0.63, respectively. Also, this study showed that cancer type was an important determinant for the deep neural network model in predicting the tumor delivery efficiency across all endpoints (19-29%). Among all physicochemical properties, the Zeta potential and core material played a greater role than other properties, such as the type, shape, and targeting strategy.

Conclusion

This study provides a quantitative model to improve the design of cancer nanomedicine with greater tumor delivery efficiency. These results help to improve our understanding of the causes of low NP tumor delivery efficiency. This study demonstrates the feasibility of integrating artificial intelligence with PBPK modeling approaches to study cancer nanomedicine.

Predicting Post-Therapeutic Visual Acuity and OCT Images in Patients With Central Serous Chorioretinopathy by Artificial Intelligence

To predict visual acuity (VA) and post-therapeutic optical coherence tomography (OCT) images 1, 3, and 6 months after laser treatment in patients with central serous chorioretinopathy (CSC) by artificial intelligence (AI). Real-world clinical and imaging data were collected at Zhongshan Ophthalmic Center (ZOC) and Xiamen Eye Center (XEC). The data obtained from ZOC (416 eyes of 401 patients) were used as the training set; the data obtained from XEC (64 eyes of 60 patients) were used as the test set. Six different machine learning algorithms and a blending algorithm were used to predict VA, and a pix2pixHD method was adopted to predict post-therapeutic OCT images in patients after laser treatment. The data for VA predictions included clinical features obtained from electronic medical records (20 features) and measured features obtained from fundus fluorescein angiography, indocyanine green angiography, and OCT (145 features). The data for OCT predictions included 480 pairs of pre- and post-therapeutic OCT images. The VA and OCT images predicted by AI were compared with the ground truth. In the VA predictions of XEC dataset, the mean absolute errors (MAEs) were 0.074–0.098 logMAR (within four to five letters), and the root mean square errors were 0.096–0.127 logMAR (within five to seven letters) for the 1-, 3-, and 6-month predictions, respectively; in the post-therapeutic OCT predictions, only about 5.15% (5 of 97) of synthetic OCT images could be accurately identified as synthetic images. The MAEs of central macular thickness of synthetic OCT images were 30.15 ± 13.28 μm and 22.46 ± 9.71 μm for the 1- and 3-month predictions, respectively. This is the first study to apply AI to predict VA and post-therapeutic OCT of patients with CSC. This work establishes a reliable method of predicting prognosis 6 months in advance; the application of AI has the potential to help reduce patient anxiety and serve as a reference for ophthalmologists when choosing optimal laser treatments.

Leveraging high-throughput screening data, deep neural networks, and conditional generative adversarial networks to advance predictive toxicology

There are currently 85,000 chemicals registered with the Environmental Protection Agency (EPA) under the Toxic Substances Control Act, but only a small fraction have measured toxicological data. To address this gap, high-throughput screening (HTS) and computational methods are vital. As part of one such HTS effort, embryonic zebrafish were used to examine a suite of morphological and mortality endpoints at six concentrations from over 1,000 unique chemicals found in the ToxCast library (phase 1 and 2). We hypothesized that by using a conditional generative adversarial network (cGAN) or deep neural networks (DNN), and leveraging this large set of toxicity data we could efficiently predict toxic outcomes of untested chemicals. Utilizing a novel method in this space, we converted the 3D structural information into a weighted set of points while retaining all information about the structure. In vivo toxicity and chemical data were used to train two neural network generators. The first was a DNN (Go-ZT) while the second utilized cGAN architecture (GAN-ZT) to train generators to produce toxicity data. Our results showed that Go-ZT significantly outperformed the cGAN, support vector machine, random forest and multilayer perceptron models in cross-validation, and when tested against an external test dataset. By combining both Go-ZT and GAN-ZT, our consensus model improved the SE, SP, PPV, and Kappa, to 71.4%, 95.9%, 71.4% and 0.673, respectively, resulting in an area under the receiver operating characteristic (AUROC) of 0.837. Considering their potential use as prescreening tools, these models could provide in vivo toxicity predictions and insight into the hundreds of thousands of untested chemicals to prioritize compounds for HT testing.

GPCR_LigandClassify.py; a rigorous machine learning classifier for GPCR targeting compounds

The current study describes the construction of various ligand-based machine learning models to be used for drug-repurposing against the family of G-Protein Coupled Receptors (GPCRs). In building these models, we collected > 500,000 data points, encompassing experimentally measured molecular association data of > 160,000 unique ligands against > 250 GPCRs. These data points were retrieved from the GPCR-Ligand Association (GLASS) database. We have used diverse molecular featurization methods to describe the input molecules. Multiple supervised ML algorithms were developed, tested and compared for their accuracy, F scores, as well as for their Matthews' correlation coefficient scores (MCC). Our data suggest that combined with molecular fingerprinting, ensemble decision trees and gradient boosted trees ML algorithms are on the accuracy border of the rather sophisticated deep neural nets (DNNs)-based algorithms. On a test dataset, these models displayed an excellent performance, reaching a ~ 90% classification accuracy. Additionally, we showcase a few examples where our models were able to identify interesting connections between known drugs from the Drug-Bank database and members of the GPCR family of receptors. Our findings are in excellent agreement with previously reported experimental observations in the literature. We hope the models presented in this paper synergize with the currently ongoing interest of applying machine learning modeling in the field of drug repurposing and computational drug discovery in general.

Learning chemistry: exploring the suitability of machine learning for the task of structure-based chemical ontology classification

Chemical data is increasingly openly available in databases such as PubChem, which contains approximately 110 million compound entries as of February 2021. With the availability of data at such scale, the burden has shifted to organisation, analysis and interpretation. Chemical ontologies provide structured classifications of chemical entities that can be used for navigation and filtering of the large chemical space. ChEBI is a prominent example of a chemical ontology, widely used in life science contexts. However, ChEBI is manually maintained and as such cannot easily scale to the full scope of public chemical data. There is a need for tools that are able to automatically classify chemical data into chemical ontologies, which can be framed as a hierarchical multi-class classification problem. In this paper we evaluate machine learning approaches for this task, comparing different learning frameworks including logistic regression, decision trees and long short-term memory artificial neural networks, and different encoding approaches for the chemical structures, including cheminformatics fingerprints and character-based encoding from chemical line notation representations. We find that classical learning approaches such as logistic regression perform well with sets of relatively specific, disjoint chemical classes, while the neural network is able to handle larger sets of overlapping classes but needs more examples per class to learn from, and is not able to make a class prediction for every molecule. Future work will explore hybrid and ensemble approaches, as well as alternative network architectures including neuro-symbolic approaches.

Predicting subretinal fluid absorption with machine learning in patients with central serous chorioretinopathy

Background

Machine learning was used to predict subretinal fluid absorption (SFA) at 1, 3 and 6 months after laser treatment in patients with central serous chorioretinopathy (CSC).

Methods

The clinical and imaging data from 480 eyes of 461 patients with CSC were collected at Zhongshan Ophthalmic Center (ZOC) and Xiamen Eye Center (XEC). The data included clinical features from electronic medical records and measured features from fundus fluorescein angiography (FFA), indocyanine green angiography (ICGA), optical coherence tomography angiography (OCTA), and optical coherence tomography (OCT). A ZOC dataset was used for training and internal validation. An XEC dataset was used for external validation. Six machine learning algorithms and a blending algorithm were trained to predict SFA in patients with CSC after laser treatment. The SFA results predicted by machine learning were compared with the actual patient prognoses. Based on the initial detailed investigation, we constructed a simplified model using fewer clinical features and OCT features for convenient application.

Results

During the internal validation, random forest performed best in SFA prediction, with accuracies of 0.651±0.068, 0.753±0.065 and 0.818±0.058 at 1, 3 and 6 months, respectively. In the external validation, XGBoost performed best at SFA prediction with accuracies of 0.734, 0.727, and 0.900 at 1, 3 and 6 months, respectively. The simplified model showed a comparable level of predictive power.

Conclusions

Machine learning can achieve high accuracy in long-term SFA predictions and identify the features relevant to CSC patients’ prognoses. Our study provides an individualized reference for ophthalmologists to treat and create a follow-up schedule for CSC patients.

Machine Learning in Predictive Toxicology: Recent Applications and Future Directions for Classification Models

In recent times, machine learning has become increasingly prominent in predictive toxicology as it has shifted from in vivo studies toward in silico studies. Currently, in vitro methods together with other computational methods such as quantitative structure-activity relationship modeling and absorption, distribution, metabolism, and excretion calculations are being used. An overview of machine learning and its applications in predictive toxicology is presented here, including support vector machines (SVMs), random forest (RF) and decision trees (DTs), neural networks, regression models, naïve Bayes, k-nearest neighbors, and ensemble learning. The recent successes of these machine learning methods in predictive toxicology are summarized, and a comparison of some models used in predictive toxicology is presented. In predictive toxicology, SVMs, RF, and DTs are the dominant machine learning methods due to the characteristics of the data available. Lastly, this review describes the current challenges facing the use of machine learning in predictive toxicology and offers insights into the possible areas of improvement in the field.

Development of In Vitro Corneal Models: Opportunity for Pharmacological Testing

The human eye is a specialized organ with a complex anatomy and physiology, because it is characterized by different cell types with specific physiological functions. Given the complexity of the eye, ocular tissues are finely organized and orchestrated. In the last few years, many in vitro models have been developed in order to meet the 3Rs principle (Replacement, Reduction and Refinement) for eye toxicity testing. This procedure is highly necessary to ensure that the risks associated with ophthalmic products meet appropriate safety criteria. In vitro preclinical testing is now a well-established practice of significant importance for evaluating the efficacy and safety of cosmetic, pharmaceutical, and nutraceutical products. Along with in vitro testing, also computational procedures, herein described, for evaluating the pharmacological profile of potential ocular drug candidates including their toxicity, are in rapid expansion. In this review, the ocular cell types and functionality are described, providing an overview about the scientific challenge for the development of three-dimensional (3D) in vitro models.

Predicting Cytotoxicity of Metal Oxide Nanoparticles using Isalos Analytics Platform

A literature curated dataset containing 24 distinct metal oxide (MexOy) nanoparticles (NPs), including 15 physicochemical, structural and assay-related descriptors, was enriched with 62 atomistic computational descriptors and exploited to produce a robust and validated in silico model for prediction of NP cytotoxicity. The model can be used to predict the cytotoxicity (cell viability) of MexOy NPs based on the colorimetric lactate dehydrogenase (LDH) assay and the luminometric adenosine triphosphate (ATP) assay, both of which quantify irreversible cell membrane damage. Out of the 77 total descriptors used, 7 were identified as being significant for induction of cytotoxicity by MexOy NPs. These were NP core size, hydrodynamic size, assay type, exposure dose, the energy of the MexOy conduction band (EC), the coordination number of the metal atoms on the NP surface (Avg. C.N. Me atoms surface) and the average force vector surface normal component of all metal atoms (v⟂ Me atoms surface). The significance and effect of these descriptors is discussed to demonstrate their direct correlation with cytotoxicity. The produced model has been made publicly available by the Horizon 2020 (H2020) NanoSolveIT project and will be added to the project's Integrated Approach to Testing and Assessment (IATA).

A practical guide to secondary pharmacology in drug discovery

Secondary pharmacological profiling is increasingly applied in pharmaceutical drug discovery to address unwanted pharmacological side effects of drug candidates before entering the clinic. Regulators, drug makers and patients share a demand for deep characterization of secondary pharmacology effects of novel drugs and their metabolites. The scope of such profiling has therefore expanded substantially in the past two decades, leading to the implementation of broad in silico profiling methods and focused in vitro off-target screening panels, to identify liabilities, but also opportunities, as early as possible. The pharmaceutical industry applies such panels at all stages of drug discovery routinely up to early development. Nevertheless, target composition, screening technologies, assay formats, interpretation and scheduling of panels can vary significantly between companies in the absence of dedicated guidelines. To contribute towards best practices in secondary pharmacology profiling, this review aims to summarize the state-of-the art in this field. Considerations are discussed with respect to panel design, screening strategy, implementation and interpretation of the data, including regulatory perspectives. The cascaded, or integrated, use of in silico and off-target profiling allows to exploit synergies for comprehensive safety assessment of drug candidates.

Predictive Models for Human Organ Toxicity Based on In Vitro Bioactivity Data and Chemical Structure

Traditional toxicity testing reliant on animal models is costly and low throughput, posing a significant challenge with the increasing numbers of chemicals that humans are exposed to in the environment. The purpose of this investigation was to build optimal prediction models for various human in vivo/organ-level toxicity end points (extracted from ChemIDPlus) using chemical structure and Tox21 in vitro quantitative high-throughput screening (qHTS) bioactivity assay data. Several supervised machine learning algorithms were applied to model 14 human toxicity end points pertaining to vascular, kidney, ureter and bladder, and liver organ systems. Three metrics were used to evaluate model performance: area under the receiver operating characteristic curve (AUC-ROC), balanced accuracy (BA), and Matthews correlation coefficient (MCC). The top four models, with AUC-ROC values >0.8, were derived for endocrine (0.90 ± 0.00), musculoskeletal (0.88 ± 0.02), peripheral nerve and sensation (0.85 ± 0.01), and brain and coverings (0.83 ± 0.02) toxicities, whereas the best model AUC-ROC values were >0.7 for the remaining 10 toxicities. Model performance was found to be dependent on the specific data set, model type, and feature selection method used. In addition, chemical structure and assay data showed different levels of contribution to the prediction of different toxicity end points. Although in vitro assay data, when combined with chemical structure, slightly improved the predictive accuracy for most end points (11 out of 14), a noteworthy finding was the near equal success of the structure-only models, which do not require Tox21 qHTS screening data, and the relatively poor performance of assay-only models. Thus, the top-performing structure-only models from this study could be applied for hazard screening of large sets of chemicals for potential human toxicity, whereas the largest assay contributions to models (i.e., cellular targets) could be used, along with the top-contributing structural features, to provide insight into toxicity mechanisms.

Prediction Model of Aryl Hydrocarbon Receptor Activation by a Novel QSAR Approach, DeepSnap-Deep Learning

The aryl hydrocarbon receptor (AhR) is a ligand-dependent transcription factor that senses environmental exogenous and endogenous ligands or xenobiotic chemicals. In particular, exposure of the liver to environmental metabolism-disrupting chemicals contributes to the development and propagation of steatosis and hepatotoxicity. However, the mechanisms for AhR-induced hepatotoxicity and tumor propagation in the liver remain to be revealed, due to the wide variety of AhR ligands. Recently, quantitative structure–activity relationship (QSAR) analysis using deep neural network (DNN) has shown superior performance for the prediction of chemical compounds. Therefore, this study proposes a novel QSAR analysis using deep learning (DL), called the DeepSnap–DL method, to construct prediction models of chemical activation of AhR. Compared with conventional machine learning (ML) techniques, such as the random forest, XGBoost, LightGBM, and CatBoost, the proposed method achieves high-performance prediction of AhR activation. Thus, the DeepSnap–DL method may be considered a useful tool for achieving high-throughput in silico evaluation of AhR-induced hepatotoxicity.

Machine Learning in Drug Discovery and Development Part 1: A Primer

Artificial intelligence, in particular machine learning (ML), has emerged as a key promising pillar to overcome the high failure rate in drug development. Here, we present a primer on the ML algorithms most commonly used in drug discovery and development. We also list possible data sources, describe good practices for ML model development and validation, and share a reproducible example. A companion article will summarize applications of ML in drug discovery, drug development, and postapproval phase.

Sperm quality of rats exposed to difenoconazole using classical parameters and surface-enhanced Raman scattering: classification performance by machine learning methods

Difenoconazole is a fungicide extensively used in agriculture. The aim of this study was to evaluate the effects of difenoconazole fungicide on the sperm quality of rats. Wistar rats were divided into four groups: control and exposed to 5 (D5), 10 (D10), or 50 mg^-1 kg bw^-1day (D50) of difenoconazole for 30 days, by gavage. Classical sperm parameters and surface-enhanced Raman scattering (SERS) were performed. Progressive motility, acrosomal integrity, and percentage of morphologically normal spermatozoa were reduced in the D10 and D50 groups in comparison with the control group. Sperm viability was reduced only in the D50 group. Sperm number in the testis and caput/corpus epididymis and daily sperm production were reduced in the three exposed groups. SERS measurements showed changes in the spectra of spermatozoa from D50 group, suggesting DNA damage. In addition, machine learning (ML) methods were used to evaluate the performance of three classification algorithms (artificial neural network-ANN, K-nearest neighbors-K-NN, and support vector machine-SVM) in the identification task of the groups exposed to difenoconazole. The results obtained by ML algorithms were very promising with accuracy ≥ 90% and validated the hypothesis of the exposure to difenoconazole reduces sperm quality. In conclusion, exposure of rats to different doses of the fungicide difenoconazole may impair sperm quality, with a recognizable classification pattern of exposure groups.

Computer-Aided Chemotaxonomy and Bioprospecting Study of Diterpenes of the Lamiaceae Family

Lamiaceae is one of the largest families of angiosperms and is classified into 12 subfamilies that are composed of 295 genera and 7775 species. It presents a variety of secondary metabolites such as diterpenes that are commonly found in their species, and some of them are known to be chemotaxonomic markers. The aim of this work was to construct a database of diterpenes and to use it to perform a chemotaxonomic analysis among the subfamilies of Lamiaceae, using molecular descriptors and self-organizing maps (SOMs). The 4115 different diterpenes corresponding to 6386 botanical occurrences, which are distributed in eight subfamilies, 66 genera, 639 different species and 4880 geographical locations, were added to SistematX. Molecular descriptors of diterpenes and their respective botanical occurrences were used to generate the SOMs. In all obtained maps, a match rate higher than 80% was observed, demonstrating a separation of the Lamiaceae subfamilies, corroborating with the morphological and molecular data proposed by Li et al. Therefore, through this chemotaxonomic study, we can predict the localization of a diterpene in a subfamily and assist in the search for secondary metabolites with specific structural characteristics, such as compounds with potential biological activity.

Comparative Study of Multitask Toxicity Modeling on a Broad Chemical Space

Acute toxicity is one of the most challenging properties to predict purely with computational methods due to its direct relationship to biological interactions. Moreover, toxicity can be represented by different end points: it can be measured for different species using different types of administration, etc., and it is questionable if the knowledge transfer between end points is possible. We performed a comparative study of prediction multitask toxicity for a broad chemical space using different descriptors and modeling algorithms and applied multitask learning for a large toxicity data set extracted from the Registry of Toxic Effects of Chemical Substances (RTECS). We demonstrated that multitask modeling provides significant improvement over single-output models and other machine learning methods. Our research reveals that multitask learning can be very useful to improve the quality of acute toxicity modeling and raises a discussion about the usage of multitask approaches for regulation purposes. Our MultiTox models are freely available in OCHEM platform ( ochem.eu/multitox ) under CC-BY-NC license.

Machine learning models for predicting endocrine disruption potential of environmental chemicals

We introduce here ML4Tox, a framework offering Deep Learning and Support Vector Machine models to predict agonist, antagonist, and binding activities of chemical compounds, in this case for the estrogen receptor ligand-binding domain. The ML4Tox models have been developed with a 10 × 5-fold cross-validation schema on the training portion of the CERAPP ToxCast dataset, formed by 1677 chemicals, each described by 777 molecular features. On the CERAPP "All Literature" evaluation set (agonist: 6319 compounds; antagonist 6539; binding 7283), ML4Tox significantly improved sensitivity over published results on all three tasks, with agonist: 0.78 vs 0.56; antagonist: 0.69 vs 0.11; binding: 0.66 vs 0.26.

Current approaches for choosing feature selection and learning algorithms in quantitative structure-activity relationships (QSAR)

INTRODUCTION

Quantitative structure-activity/property relationships (QSAR/QSPR) are statistical models which quantitatively correlate quantitative chemical structure information (described as molecular descriptors) to the response end points (biological activity, property, toxicity, etc.). Important strategies for QSAR model development and validation include dataset curation, variable selection, and dataset division, selection of modeling algorithms and appropriate measures of model validation. Areas covered: Different feature selection methods and various linear and nonlinear learning algorithms are employed to address the complexity of data sets for selection of appropriate features important for the responses being modeled, to reduce overfitting of the models, and to derive interpretable models. This review provides an overview of various feature selection methods as well as different statistical learning algorithms for QSAR modeling at an elementary level for nonexpert readers. Expert opinion: Novel sets of descriptors are being continuously introduced to this field; therefore, to handle this issue, there is a need to improve new tools for feature selection, which can lead to development of statistically meaningful models, usable by nonexperts in the fields. While handling data sets of limited size, special techniques like double cross-validation and consensus modeling might be more meaningful in order to remove the possibility of bias in descriptor selection.

G-Networks to Predict the Outcome of Sensing of Toxicity

G-Networks and their simplified version known as the Random Neural Network have often been used to classify data. In this paper, we present a use of the Random Neural Network to the early detection of potential of toxicity chemical compounds through the prediction of their bioactivity from the compounds' physico-chemical structure, and propose that it be automated using machine learning (ML) techniques. Specifically the Random Neural Network is shown to be an effective analytical tool to this effect, and the approach is illustrated and compared with several ML techniques.