LeadScope: software for exploring large sets of screening data

Improving predictions and understanding of primary and ultimate biodegradation rates with machine learning models

This study aimed to develop machine learning based quantitative structure biodegradability relationship (QSBR) models for predicting primary and ultimate biodegradation rates of organic chemicals, which are essential parameters for environmental risk assessment. For this purpose, experimental primary and ultimate biodegradation rates of high consistency were compiled for 173 organic compounds. A significant number of descriptors were calculated with a collection of quantum/computational chemistry software and tools to achieve comprehensive representation and interpretability. Following a pre-screening process, multiple QSBR models were developed for both primary and ultimate endpoints using three algorithms: extreme gradient boosting (XGBoost), support vector machine (SVM), and multiple linear regression (MLR). Furthermore, a unified QSBR model was constructed using the knowledge transfer technique and XGBoost. Results demonstrated that all QSBR models developed in this study had good performance. Particularly, SVM models exhibited high level of goodness of fit (coefficient of determination on the training set of 0.973 for primary and 0.980 for ultimate), robustness (leave-one-out cross-validated coefficient of 0.953 for primary and 0.967 for ultimate), and external predictive ability (external explained variance of 0.947 for primary and 0.958 for ultimate). The knowledge transfer technique enhanced model performance by learning from properties of two biodegradation endpoints. Williams plots were used to visualize the application domains of the models. Through SHapley Additive exPlanations (SHAP) analysis, this study identified key features affecting biodegradation rates. Notably, MDEO-12, APC2D1_C_O, and other features contributed to primary biodegradation, while AATS0v, AATS2v, and others inhibited it. For ultimate biodegradation, features like No. of Rotatable Bonds, APC2D1_C_O, and minHBa were contributors, while C1SP3, Halogen Ratio, GGI4, and others hindered the process. Also, the study quantified the contributions of each feature in predictions for individual chemicals. This research provides valuable tools for predicting both primary and ultimate biodegradation rates while offering insights into the mechanisms.

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.

A method for in vitro data and structure curation to optimize for QSAR modelling of minimum absolute potency levels and a comparative use case

Large screening programs such as the US Tox21 are releasing experimental in vitro results for many endpoints of relevance for human health. In (Q)SAR modelling, it is essential to clearly define the endpoint (OECD QSAR Validation Principle 1) and extract the most robust data points according to the definition. We have developed a comprehensive data curation procedure to interpret in vitro experimental data sets for (Q)SAR development, with modules for selecting actives according to quality of curve fittings, magnitude of activity and 'absolute' potency cut-offs, requiring non-cytotoxicity at activity concentration; extracting only very robust inactives; selecting only substances tested in high purity; and accounting for assay signal interference. A structure curation procedure with uniform representation of tautomeric classes of substances is also developed. The detailed method and a use case of modelling Tox21 data for an estrogen receptor α agonism assay with and without use of the method is presented.

Development of QSAR models to predict blood-brain barrier permeability

Assessing drug permeability across the blood-brain barrier (BBB) is important when evaluating the abuse potential of new pharmaceuticals as well as developing novel therapeutics that target central nervous system disorders. One of the gold-standard in vivo methods for determining BBB permeability is rodent log BB; however, like most in vivo methods, it is time-consuming and expensive. In the present study, two statistical-based quantitative structure-activity relationship (QSAR) models were developed to predict BBB permeability of drugs based on their chemical structure. The in vivo BBB permeability data were harvested for 921 compounds from publicly available literature, non-proprietary drug approval packages, and University of Washington's Drug Interaction Database. The cross-validation performance statistics for the BBB models ranged from 82 to 85% in sensitivity and 80-83% in negative predictivity. Additionally, the performance of newly developed models was assessed using an external validation set comprised of 83 chemicals. Overall, performance of individual models ranged from 70 to 75% in sensitivity, 70-72% in negative predictivity, and 78-86% in coverage. The predictive performance was further improved to 93% in coverage by combining predictions across the two software programs. These new models can be rapidly deployed to predict blood brain barrier permeability of pharmaceutical candidates and reduce the use of experimental animals.

Novel machine learning models to predict endocrine disruption activity for high-throughput chemical screening

An area of ongoing concern in toxicology and chemical risk assessment is endocrine disrupting chemicals (EDCs). However, thousands of legacy chemicals lack the toxicity testing required to assess their respective EDC potential, and this is where computational toxicology can play a crucial role. The US (United States) Environmental Protection Agency (EPA) has run two programs, the Collaborative Estrogen Receptor Activity Project (CERAPP) and the Collaborative Modeling Project for Receptor Activity (CoMPARA) which aim to predict estrogen and androgen activity, respectively. The US EPA solicited research groups from around the world to provide endocrine receptor activity Qualitative (or Quantitative) Structure Activity Relationship ([Q]SAR) models and then combined them to create consensus models for different toxicity endpoints. Random Forest (RF) models were developed to cover a broader range of substances with high predictive capabilities using large datasets from CERAPP and CoMPARA for estrogen and androgen activity, respectively. By utilizing simple descriptors from open-source software and large training datasets, RF models were created to expand the domain of applicability for predicting endocrine disrupting activity and help in the screening and prioritization of extensive chemical inventories. In addition, RFs were trained to conservatively predict the activity, meaning models are more likely to make false-positive predictions to minimize the number of False Negatives. This work presents twelve binary and multi-class RF models to predict binding, agonism, and antagonism for estrogen and androgen receptors. The RF models were found to have high predictive capabilities compared to other in silico modes, with some models reaching balanced accuracies of 93% while having coverage of 89%. These models are intended to be incorporated into evolving priority-setting workflows and integrated strategies to support the screening and selection of chemicals for further testing and assessment by identifying potential endocrine-disrupting substances.

Nonclinical safety evaluation of a novel ionizable lipid for mRNA delivery

mRNA vaccines hold tremendous potential in disease control and prevention for their flexibility with respect to production, application, and design. Recent breakthroughs in mRNA vaccination would have not been possible without major advances in lipid nanoparticles (LNPs) technologies. We developed an LNP containing a novel ionizable cationic lipid, Lipid-1, and three well known excipients. An in silico toxicity hazard assessment for genotoxicity, a genotoxicity assessment, and a dose range finding toxicity study were performed to characterize the safety profile of Lipid-1. The in silico toxicity hazard assessment, utilizing two prediction systems DEREK and Leadscope, did not find any structural alert for mutagenicity and clastogenicity, and prediction in the statistical models were all negative. In addition, applying a read-across approach a structurally very similar compound was tested negative in two in vitro assays confirming the low genotoxicity potential of Lipid-1. A dose range finding toxicity study in rabbits, receiving a single intramuscular injection of either different doses of an mRNA encoding Influenza Hemagglutinin H3 antigen encapsulated in the LNP containing Lipid-1 or the empty LNP, evaluated local tolerance and systemic toxicity during a 2-week observation period. Only rabbits exposed to the vaccine were able to develop a specific IgG response, indicating an appropriate vaccine take. The vaccine was well tolerated up to 250 μg mRNA/injection, which was defined as the No Observed Adverse Effect Level (NOAEL). These results support the use of the LNP containing Lipid-1 as an mRNA delivery system for different vaccine formulations and its deployment into clinical trials.

The VEGAHUB Platform: The Philosophy and the Tools

VEGAHUB (www.vegahub.eu) is a repository of freely available, downloadable tools based on computational toxicology methodologies. The main software tool available in VEGAHUB is VEGA QSAR software encoding more than 90 quantitative structure-activity relationship (QSAR) models for tens of endpoints for human toxicology, ecotoxicology, environmental, physico-chemical and toxicokinetic properties. However, beyond VEGA QSAR, VEGAHUB offers several other tools. Here, we present these resources, the possibilities to fully exploit them and the ways in which to integrate results provided by different VEGAHUB tools. Read-across and weight-of-evidence represent a major advantage of VEGAHUB. Integration between hazard and exposure is provided within innovative tools, which are specific for well-defined scenarios, such as those for cosmetic products. Prioritisation can be achieved by integrating results from 48 models. Finally, we highlight how some tools may not only fit predefined endpoints but also could be applied to general problems and research applications in the QSAR field. A couple of examples are provided, in which a critical assessment of the predictions and the documentation associated with the prediction are considered, in order to properly assess the quality of the results. These results may be associated with different levels of uncertainty or even be conflicting.

Implementation of in silico toxicology protocols within a visual and interactive hazard assessment platform

Mechanistically-driven alternative approaches to hazard assessment invariably require a battery of tests, including both in silico models and experimental data. The decision-making process, from selection of the methods to combining the information based on the weight-of-evidence, is ideally described in published guidelines or protocols. This ensures that the application of such approaches is defendable to reviewers within regulatory agencies and across the industry. Examples include the ICH M7 pharmaceutical impurities guideline and the published in silico toxicology protocols. To support an efficient, transparent, consistent and fully documented implementation of these protocols, a new and novel interactive software solution is described to perform such an integrated hazard assessment based on public and proprietary information.

In Silico Models for Predicting Acute Systemic Toxicity

In this chapter, we give a brief overview of the regulatory requirements for acute systemic toxicity information in the European Union, and we review structure-based computational models that are available and potentially useful in the assessment of acute systemic toxicity. Emphasis is placed on quantitative structure-activity relationship (QSAR) models implemented by means of a range of software tools. The most recently published literature models for acute systemic toxicity are also discussed, and perspectives for future developments in this field are offered.

In Silico Models for Skin Sensitization and Irritation

The assessment of skin irritation, and in particular of skin sensitization, has undergone an evolution process over the last years, pushing forward to new heights of quality and innovation. Public and commercial in silico tools have been developed for skin sensitization and irritation, introducing the possibility to simplify the evaluation process and the development of topical products within the dogma of the computational methods, representing the new doctrine in the field of risk assessment.The possibility of using in silico methods is particularly appealing and advantageous due to their high speed and low-cost results.The most widespread and popular topical products are represented by cosmetics. The European Regulation 1223/2009 on cosmetic products represents a paradigm shift for the safety assessment of cosmetics transitioning from a classical toxicological approach based on animal testing, towards a completely novel strategy, where the use of animals for toxicity testing is completely banned. In this context sustainable alternatives to animal testing need to be developed, especially for skin sensitization and irritation, two critical and fundamental endpoints for the assessment of cosmetics.The Quantitative Risk Assessment (QRA) methodology and the risk assessment using New Approach Methodologies (NAM) represent new frontiers to further improve the risk assessment process for these endpoints, in particular for skin sensitization.In this chapter we present an overview of the already existing models for skin sensitization and irritation. Some examples are presented here to illustrate tools and platforms used for the evaluation of chemicals.

Designing Out PXR Activity on Drug Discovery Projects: A Review of Structure-Based Methods, Empirical and Computational Approaches

This perspective discusses the role of pregnane xenobiotic receptor (PXR) in drug discovery and the impact of its activation on CYP3A4 induction. The use of structural biology to reduce PXR activity on drug discovery projects has become more common in recent years. Analysis of this work highlights several important molecular interactions, and the resultant structural modifications to reduce PXR activity are summarized. The computational approaches undertaken to support the design of new drugs devoid of PXR activation potential are also discussed. Finally, the SAR of empirical design strategies to reduce PXR activity is reviewed, and the key SAR transformations are discussed and summarized. In conclusion, this perspective demonstrates that PXR activity can be greatly diminished or negated on active drug discovery projects with the knowledge now available. This perspective should be useful to anyone who seeks to reduce PXR activity on a drug discovery project.

CATMoS: Collaborative Acute Toxicity Modeling Suite

BACKGROUND

Humans are exposed to tens of thousands of chemical substances that need to be assessed for their potential toxicity. Acute systemic toxicity testing serves as the basis for regulatory hazard classification, labeling, and risk management. However, it is cost- and time-prohibitive to evaluate all new and existing chemicals using traditional rodent acute toxicity tests. In silico models built using existing data facilitate rapid acute toxicity predictions without using animals.

OBJECTIVES

The U.S. Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) Acute Toxicity Workgroup organized an international collaboration to develop in silico models for predicting acute oral toxicity based on five different end points: Lethal Dose 50 (LD50 value, U.S. Environmental Protection Agency hazard (four) categories, Globally Harmonized System for Classification and Labeling hazard (five) categories, very toxic chemicals [LD50 (LD50≤50mg/kg)], and nontoxic chemicals (LD50>2,000mg/kg).

METHODS

An acute oral toxicity data inventory for 11,992 chemicals was compiled, split into training and evaluation sets, and made available to 35 participating international research groups that submitted a total of 139 predictive models. Predictions that fell within the applicability domains of the submitted models were evaluated using external validation sets. These were then combined into consensus models to leverage strengths of individual approaches.

RESULTS

The resulting consensus predictions, which leverage the collective strengths of each individual model, form the Collaborative Acute Toxicity Modeling Suite (CATMoS). CATMoS demonstrated high performance in terms of accuracy and robustness when compared with in vivo results.

DISCUSSION

CATMoS is being evaluated by regulatory agencies for its utility and applicability as a potential replacement for in vivo rat acute oral toxicity studies. CATMoS predictions for more than 800,000 chemicals have been made available via the National Toxicology Program's Integrated Chemical Environment tools and data sets (ice.ntp.niehs.nih.gov). The models are also implemented in a free, standalone, open-source tool, OPERA, which allows predictions of new and untested chemicals to be made. https://doi.org/10.1289/EHP8495.

The performance, reliability and potential application of in silico models for predicting the acute oral toxicity of pharmaceutical compounds

Acute oral toxicity (AOT) information is utilized to categorize compounds according to the severity of their hazard and used to inform risk assessments for human health and the environment. Given the wealth of historical AOT information and technological advances, in silico models are being created and evaluated as potential tools to predict the AOT of compounds and reduce reliance on animal testing. Utilizing a historical database of AOT data on 371 Bristol Myers Squibb pharmaceutical compounds (PCs) (195 pharmaceutical intermediates and 176 active pharmaceutical ingredients), we evaluated two pioneering in silico AOT programs: the Leadscope Acute Oral Toxicity Model Suite and the Collaborative Acute Toxicity Modeling Suite. These models demonstrated a high degree of agreement with the in vivo results as well as a high level of sensitivity. We found that these models can be effectively utilized to identify PCs which are of low acute oral toxicity (LD₅₀ > 2000 mg/kg), PCs which should not be classified as Dangerous Goods (LD₅₀ > 300 mg/kg), and can assist in identifying a starting dose for in vivo AOT studies. This manuscript provides an evaluation of the performance of these in silico models and proposes use cases where these in silico models can be most confidently and effectively employed.

Current development of integrated web servers for preclinical safety and pharmacokinetics assessments in drug development

In drug development, preclinical safety and pharmacokinetics assessments of candidate drugs to ensure the safety profile are a must. While in vivo and in vitro tests are traditionally used, experimental determinations have disadvantages, as they are usually time-consuming and costly. In silico predictions of these preclinical endpoints have each been developed in the past decades. However, only a few web-based tools have integrated different models to provide a simple one-step platform to help researchers thoroughly evaluate potential drug candidates. To efficiently achieve this approach, a platform for preclinical evaluation must not only predict key ADMET (absorption, distribution, metabolism, excretion and toxicity) properties but also provide some guidance on structural modifications to improve the undesired properties. In this review, we organized and compared several existing integrated web servers that can be adopted in preclinical drug development projects to evaluate the subject of interest. We also introduced our new web server, Virtual Rat, as an alternative choice to profile the properties of drug candidates. In Virtual Rat, we provide not only predictions of important ADMET properties but also possible reasons as to why the model made those structural predictions. Multiple models were implemented into Virtual Rat, including models for predicting human ether-a-go-go-related gene (hERG) inhibition, cytochrome P450 (CYP) inhibition, mutagenicity (Ames test), blood-brain barrier penetration, cytotoxicity and Caco-2 permeability. Virtual Rat is free and has been made publicly available at https://virtualrat.cmdm.tw/.

Development of improved QSAR models for predicting the outcome of the in vivo micronucleus genetic toxicity assay

All drugs entering clinical trials are expected to undergo a series of in vitro and in vivo genotoxicity tests as outlined in the International Council on Harmonization (ICH) S2 (R1) guidance. Among the standard battery of genotoxicity tests used for pharmaceuticals, the in vivo micronucleus assay, which measures the frequency of micronucleated cells mostly from blood or bone marrow, is recommended for detecting clastogens and aneugens. (Quantitative) structure-activity relationship [(Q)SAR] models may be used as early screening tools by pharmaceutical companies to assess genetic toxicity risk during drug candidate selection. Models can also provide decision support information during regulatory review as part of the weight-of-evidence when experimental data are insufficient. In the present study, two commercial (Q)SAR platforms were used to construct in vivo micronucleus models from a recently enhanced in-house database of non-proprietary study findings in mice. Cross-validated performance statistics for the new models showed sensitivity of up to 74% and negative predictivity of up to 86%. In addition, the models demonstrated cross-validated specificity of up to 77% and coverage of up to 94%. These new models will provide more reliable predictions and offer an investigational approach for drug safety assessment with regards to identifying potentially genotoxic compounds.

Structural Alerts and Random Forest Models in a Consensus Approach for Receptor Binding Molecular Initiating Events

A molecular initiating event (MIE) is the gateway to an adverse outcome pathway (AOP), a sequence of events ending in an adverse effect. In silico predictions of MIEs are a vital tool in a modern, mechanism-focused approach to chemical risk assessment. For 90 biological targets representing important human MIEs, structural alert-based models have been constructed with an automated procedure that uses Bayesian statistics to iteratively select substructures. These models give impressive average performance statistics (an average of 92% correct predictions across targets), significantly improving on previous models. Random Forest models have been constructed from physicochemical features for the same targets, giving similarly impressive performance statistics (93% correct predictions). A key difference between the models is interpretation of predictions-the structural alert models are transparent and easy to interpret, while Random Forest models can only identify the most important physicochemical features for making predictions. The two complementary models have been combined in a consensus model, improving performance compared to each individual model (94% correct predictions) and increasing confidence in predictions. Variation in model performance has been explained by calculating a modelability index (MODI), using Tanimoto coefficient between Morgan fingerprints to identify nearest neighbor chemicals. This work is an important step toward building confidence in the use of in silico tools for assessment of toxicity.

Leveraging complementary computational models for prioritizing chemicals of developmental and reproductive toxicity concern: an example of food contact materials

The evaluation of developmental and reproductive toxicity of food contact materials (FCMs) is an important task for food safety. Since traditional experiments are both time-consuming and labor-intensive, only a small number of FCMs have sufficient toxicological data for evaluating their effects on human health. While computational methods such as structural alerts and quantitative structure-activity relationships can serve as first-line tools for the identification of chemicals of high toxicity concern, models with binary outputs and unsatisfied accuracy and coverage prevent the use of computational methods for prioritizing chemicals of high concern. This study proposed a genetic algorithm-based method to develop a weight-of-evidence (WoE) model leveraging complementary methods of structural alerts, quantitative structure-activity relationships and in silico toxicogenomics models for chemical prioritization. The WoE model was applied to evaluate 623 food contact chemicals and identify 26 chemicals of high toxicity concern, where 13 chemicals have been reported to be developmental or reproductive toxic and further experiments are suggested for the remaining 13 chemicals without toxicity data related to developmental and reproductive effects. The proposed WoE model is potentially useful for prioritizing chemicals of high toxicity concern and the methodology may be applied to toxicities other than developmental and reproductive toxicity.

Transitioning to composite bacterial mutagenicity models in ICH M7 (Q)SAR analyses

The International Council on Harmonisation (ICH) M7(R1) guideline describes the use of complementary (quantitative) structure-activity relationship ((Q)SAR) models to assess the mutagenic potential of drug impurities in new and generic drugs. Historically, the CASE Ultra and Leadscope software platforms used two different statistical-based models to predict mutations at G-C (guanine-cytosine) and A-T (adenine-thymine) sites, to comprehensively assess bacterial mutagenesis. In the present study, composite bacterial mutagenicity models covering multiple mutation types were developed. These new models contain more than double the number of chemicals (n = 9,254 and n = 13,514) than the corresponding non-composite models and show better toxicophore coverage. Additionally, the use of a single composite bacterial mutagenicity model simplifies impurity analysis in an ICH M7 (Q)SAR workflow by reducing the number of model outputs requiring review. An external validation set of 388 drug impurities representing proprietary pharmaceutical chemical space showed performance statistics ranging from of 66-82% in sensitivity, 91-95% in negative predictivity and 96% in coverage. This effort represents a major enhancement to these (Q)SAR models and their use under ICH M7(R1), leading to improved patient safety through greater predictive accuracy, applicability, and efficiency when assessing the bacterial mutagenic potential of drug impurities.

Reaching for the bright StARs in chemical space

Visualization of activity data in chemical space is common in drug discovery. Navigating the space in a systematic manner is not trivial, given its size and huge coverage. To this end, methods for data visualization have been developed charting biological activity into chemical space. Herein, we review the progress in different visualization approaches to explore the chemical space aiming at reaching insightful structure-activity relationships (SARs) in the chemical space. We discuss recent methods including consensus diversity plots, ChemMaps, and constellation plots. Several of the methods we review can be extended to analyze other properties of interest in medicinal chemistry, such as structure-toxicity relationships, and can be adapted to postprocess results of virtual screening (VS) of large compound libraries.

QSAR modelling of a large imbalanced aryl hydrocarbon activation dataset by rational and random sampling and screening of 80,086 REACH pre-registered and/or registered substances

The Aryl hydrocarbon receptor (AhR) plays important roles in many normal and pathological physiological processes, including endocrine homeostasis, foetal development, cell cycle regulation, cellular oxidation/antioxidation, immune regulation, metabolism of endogenous and exogenous substances, and carcinogenesis. An experimental data set for human in vitro AhR activation comprising 324,858 substances, of which 1,982 were confirmed actives, was used to test an in-house-developed approach to rationally select Quantitative Structure-Activity Relationship (QSAR) training set substances from an unbalanced data set. In the first iteration, active and inactive substances were selected by random to make QSAR models. Then, more inactive substances were added to the training set in two further iterations based on incorrect or out-of-domain predictions to produce larger models. The resulting ‘rational’ model, comprising 832 actives and four times as many inactives, i.e. 3,328, was compared to a model with a training set of same size and proportion of inactives chosen entirely by random. Both models underwent robust cross-validation and external validation showing good statistical performance, with the rational model having external validation sensitivity of 85.1% and specificity of 97.1%, compared to the random model with sensitivity 89.1% and specificity 91.3%. Furthermore, we integrated the training sets for both models with the 93 external validation test set actives and 372 randomly selected inactives to make two final models. They also underwent external validations for specificity and cross-validations, which confirmed that good predictivity was maintained. All developed models were applied to predict 80,086 EU REACH substances. The rational and random final models had 63.1% and 56.9% coverage of the REACH set, respectively, and predicted 1,256 and 3,214 substances as actives. The final models as well as predictions for AhR activation for 650,000 substances will be published in the Danish (Q)SAR Database and can, for example, be used for priority setting, in read-across predictions and in weight-of-evidence assessments of chemicals.

Computational prediction models for assessing endocrine disrupting potential of chemicals

Endocrine disrupting chemicals (EDCs) mimic natural hormones and disrupt endocrine function. Humans and wildlife are exposed to EDCs might alter endocrine functions through various mechanisms and lead to an adverse effects. Hence, EDCs identification is important to protect the ecosystem and to promote the public health. Leveraging in-vitro and in-vivo experiments to identify potential EDCs is time consuming and expensive. Hence, quantitative structure-activity relationship is applied to screen the potential EDCs. Here, we summarize the predictive models developed using various algorithms to forecast the binding activity of chemicals to the estrogen and androgen receptors, alpha-fetoprotein, and sex hormone binding globulin.

Predicting Aromatic Amine Mutagenicity with Confidence: A Case Study Using Conformal Prediction

The occurrence of mutagenicity in primary aromatic amines has been investigated using conformal prediction. The results of the investigation show that it is possible to develop mathematically proven valid models using conformal prediction and that the existence of uncertain classes of prediction, such as both (both classes assigned to a compound) and empty (no class assigned to a compound), provides the user with additional information on how to use, further develop, and possibly improve future models. The study also indicates that the use of different sets of fingerprints results in models, for which the ability to discriminate varies with respect to the set level of acceptable errors.

Combination of experimental and in silico methods for the assessment of the phototransformation products of the antipsychotic drug/metabolite Mesoridazine

The lack of studies on the fate and effects of drug metabolites in the environment is of concern. As their parent compounds, metabolites enter the aquatic environment and are subject to biotic and abiotic process. In this regard, photolysis plays an important role. This study combined experimental and in silico quantitative structure-activity relationship (QSAR) methods to assess the fate and effects of Mesoridazine (MESO), a pharmacologically active human drug and metabolite of the antipsychotic agent Thioridazine, and its transformation products (TPs) formed through a Xenon lamp irradiation. After 256min, the photodegradation of MESO⋅besylate (50mgL^-1) achieved 90.4% and 6.9% of primary elimination and mineralization, respectively. The photon flux emitted by the lamp (200-600nm) was 169.55Jcm^-2. Sixteen TPs were detected by means of liquid chromatography-high resolution mass spectrometry (LC-HRMS), and the structures were proposed based on MSⁿ fragmentation patterns. The main transformation reactions were sulfoxidation, hydroxylation, dehydrogenation, and sulfoxide elimination. A back-transformation of MESO to Thioridazine was evidenced. Aerobic biodegradation tests (OECD 301 D and 301F) were applied to MESO and the mixture of TPs present after 256min of photolysis. Most of TPs were not biodegraded, demonstrating their tendency to persist in aquatic environments. The ecotoxicity towards Vibrio fischeri showed a decrease in toxicity during the photolysis process. The in silico QSAR tools QSARINS and US-EPA PBT profiler were applied for the screening of TPs with character of persistence, bioaccumulation, and toxicity (PBT). They have revealed the carbazole derivatives TP 355 and TP 337 as PBT/vPvB (very persistent and very bioaccumulative) compounds. In silico QSAR predictions for mutagenicity and genotoxicity provided by CASE Ultra and Leadscope® indicated positive alerts for mutagenicity on TP 355 and TP 337. Further studies regarding the carbazole derivative TPs should be considered to confirm their hazardous character.

Irreversible inhibition of cytosolic thioredoxin reductase 1 as a mechanistic basis for anticancer therapy

Cancer cells adapt to their inherently increased oxidative stress through activation of the glutathione (GSH) and thioredoxin (TXN) systems. Inhibition of both of these systems effectively kills cancer cells, but such broad inhibition of antioxidant activity also kills normal cells, which is highly unwanted in a clinical setting. We therefore evaluated targeting of the TXN pathway alone and, more specifically, selective inhibition of the cytosolic selenocysteine-containing enzyme TXN reductase 1 (TXNRD1). TXNRD1 inhibitors were discovered in a large screening effort and displayed increased specificity compared to pan-TXNRD inhibitors, such as auranofin, that also inhibit the mitochondrial enzyme TXNRD2 and additional targets. For our lead compounds, TXNRD1 inhibition correlated with cancer cell cytotoxicity, and inhibitor-triggered conversion of TXNRD1 from an antioxidant to a pro-oxidant enzyme correlated with corresponding increases in cellular production of H2O2 In mice, the most specific TXNRD1 inhibitor, here described as TXNRD1 inhibitor 1 (TRi-1), impaired growth and viability of human tumor xenografts and syngeneic mouse tumors while having little mitochondrial toxicity and being better tolerated than auranofin. These results display the therapeutic anticancer potential of irreversibly targeting cytosolic TXNRD1 using small molecules and present potent and selective TXNRD1 inhibitors. Given the pronounced up-regulation of TXNRD1 in several metastatic malignancies, it seems worthwhile to further explore the potential benefit of specific irreversible TXNRD1 inhibitors for anticancer therapy.

Chemical structure and correlation analysis of HIV-1 NNRT and NRT inhibitors and database-curated, published inhibition constants with chemical structure in diverse datasets

Human immunodeficiency virus (HIV-1) reverse transcriptase is a major target for designing anti-HIV drugs. Developed inhibitors are divided into non-nucleoside analog reverse-transcriptase inhibitors (NNRTIs) and nucleoside analog reverse-transcriptase inhibitors (NRTIs) depending on their mechanism. Given that many inhibitors have been studied and for many of them binding affinity constants have been calculated, it is beneficial to analyze the chemical landscape of these families of inhibitors and correlate these inhibition constants with molecular structure descriptors. For this, the HIV-1 RT data was retrieved from the ChEMBL database, carefully curated, and original literature verified, grouped into NRTIs and NNRTIs, analyzed using a hierarchical scaffold classification method and modelled with best multi-linear regression approach. Analysis of the HIV-1 NNRTIs subset results in ten different common structural parent types of oxazepanone, piperazinone, pyrazine, oxazinanone, diazinanone, pyridine, pyrrole, diazepanone, thiazole, and triazine. The same analysis for HIV-1 NRTIs groups structures into four different parent types of uracil, pyrimide, pyrimidione, and imidazole. Each scaffold tree corresponding to the parent types has been carefully analyzed and examined, and changes in chemical structure favorable to potency and stability are highlighted. For both subsets, descriptive and predictive QSAR models are derived, discussed and externally validated, revealing general trends in relationships between molecular structure and binding affinity constants in structurally diverse datasets. Data and QSAR models are available at the QsarDB repository (http://dx.doi.org/10.15152/QDB.202).

An ensemble model of QSAR tools for regulatory risk assessment

Quantitative structure activity relationships (QSARs) are theoretical models that relate a quantitative measure of chemical structure to a physical property or a biological effect. QSAR predictions can be used for chemical risk assessment for protection of human and environmental health, which makes them interesting to regulators, especially in the absence of experimental data. For compatibility with regulatory use, QSAR models should be transparent, reproducible and optimized to minimize the number of false negatives. In silico QSAR tools are gaining wide acceptance as a faster alternative to otherwise time-consuming clinical and animal testing methods. However, different QSAR tools often make conflicting predictions for a given chemical and may also vary in their predictive performance across different chemical datasets. In a regulatory context, conflicting predictions raise interpretation, validation and adequacy concerns. To address these concerns, ensemble learning techniques in the machine learning paradigm can be used to integrate predictions from multiple tools. By leveraging various underlying QSAR algorithms and training datasets, the resulting consensus prediction should yield better overall predictive ability. We present a novel ensemble QSAR model using Bayesian classification. The model allows for varying a cut-off parameter that allows for a selection in the desirable trade-off between model sensitivity and specificity. The predictive performance of the ensemble model is compared with four in silico tools (Toxtree, Lazar, OECD Toolbox, and Danish QSAR) to predict carcinogenicity for a dataset of air toxins (332 chemicals) and a subset of the gold carcinogenic potency database (480 chemicals). Leave-one-out cross validation results show that the ensemble model achieves the best trade-off between sensitivity and specificity (accuracy: 83.8 % and 80.4 %, and balanced accuracy: 80.6 % and 80.8 %) and highest inter-rater agreement [kappa (κ): 0.63 and 0.62] for both the datasets. The ROC curves demonstrate the utility of the cut-off feature in the predictive ability of the ensemble model. This feature provides an additional control to the regulators in grading a chemical based on the severity of the toxic endpoint under study.

CERAPP: Collaborative Estrogen Receptor Activity Prediction Project

BACKGROUND

Humans are exposed to thousands of man-made chemicals in the environment. Some chemicals mimic natural endocrine hormones and, thus, have the potential to be endocrine disruptors. Most of these chemicals have never been tested for their ability to interact with the estrogen receptor (ER). Risk assessors need tools to prioritize chemicals for evaluation in costly in vivo tests, for instance, within the U.S. EPA Endocrine Disruptor Screening Program.

OBJECTIVES

We describe a large-scale modeling project called CERAPP (Collaborative Estrogen Receptor Activity Prediction Project) and demonstrate the efficacy of using predictive computational models trained on high-throughput screening data to evaluate thousands of chemicals for ER-related activity and prioritize them for further testing.

METHODS

CERAPP combined multiple models developed in collaboration with 17 groups in the United States and Europe to predict ER activity of a common set of 32,464 chemical structures. Quantitative structure-activity relationship models and docking approaches were employed, mostly using a common training set of 1,677 chemical structures provided by the U.S. EPA, to build a total of 40 categorical and 8 continuous models for binding, agonist, and antagonist ER activity. All predictions were evaluated on a set of 7,522 chemicals curated from the literature. To overcome the limitations of single models, a consensus was built by weighting models on scores based on their evaluated accuracies.

RESULTS

Individual model scores ranged from 0.69 to 0.85, showing high prediction reliabilities. Out of the 32,464 chemicals, the consensus model predicted 4,001 chemicals (12.3%) as high priority actives and 6,742 potential actives (20.8%) to be considered for further testing.

CONCLUSION

This project demonstrated the possibility to screen large libraries of chemicals using a consensus of different in silico approaches. This concept will be applied in future projects related to other end points.

CITATION

Mansouri K, Abdelaziz A, Rybacka A, Roncaglioni A, Tropsha A, Varnek A, Zakharov A, Worth A, Richard AM, Grulke CM, Trisciuzzi D, Fourches D, Horvath D, Benfenati E, Muratov E, Wedebye EB, Grisoni F, Mangiatordi GF, Incisivo GM, Hong H, Ng HW, Tetko IV, Balabin I, Kancherla J, Shen J, Burton J, Nicklaus M, Cassotti M, Nikolov NG, Nicolotti O, Andersson PL, Zang Q, Politi R, Beger RD, Todeschini R, Huang R, Farag S, Rosenberg SA, Slavov S, Hu X, Judson RS. 2016.

CERAPP

Collaborative Estrogen Receptor Activity Prediction Project. Environ Health Perspect 124:1023-1033; http://dx.doi.org/10.1289/ehp.1510267.

In Silico Models for Acute Systemic Toxicity

In this chapter, we give an overview of the regulatory requirements for acute systemic toxicity information in the European Union, and we review the availability of structure-based computational models that are available and potentially useful in the assessment of acute systemic toxicity. The most recently published literature models for acute systemic toxicity are also discussed, and perspectives for future developments in this field are offered.

Environmental fate and effect assessment of thioridazine and its transformation products formed by photodegradation

An experimental and in silico quantitative structure-activity relationship (QSAR) approach was applied to assess the environmental fate and effects of the antipsychotic drug Thioridazine (THI). The sunlight-driven attenuation of THI was simulated using a Xenon arc lamp. The photodegradation reached the complete primary elimination, whereas 97% of primary elimination and 11% of mineralization was achieved after 256 min of irradiation for the initial concentrations of 500 μg L(-1) and 50 mg L(-1), respectively. A non-target approach for the identification and monitoring of transformation products (TPs) was adopted. The structure of the TPs was further elucidated using liquid chromatography-high resolution mass spectrometry (LC-HRMS). The proposed photodegradation pathway included sulfoxidation, hydroxylation, dehydroxylation, and S- and N-dealkylation, taking into account direct and indirect photolysis through a self-sensitizing process in the higher concentration studied. The biodegradability of THI and photolytic samples of THI was tested according to OECD 301D and 301F, showing that THI and the mixture of TPs were not readily biodegradable. Furthermore, THI was shown to be highly toxic to environmental bacteria using a modified luminescent bacteria test with Vibrio fischeri. This bacteriotoxic activity of THI was significantly reduced by phototransformation and individual concentration-response analysis confirmed a lowered bacterial toxicity for the sulfoxidation products Thioridazine-2-sulfoxide and Thioridazine-5-sulfoxide. Additionally, the applied QSAR models predicted statistical and rule-based positive alerts of mutagenic activities for carbazole derivative TPs (TP 355 and TP 339) formed through sulfoxide elimination, which would require further confirmatory in vitro validation tests.

Initial hazard screening for genotoxicity of photo-transformation products of ciprofloxacin by applying a combination of experimental and in-silico testing

Ciprofloxacin (CIP) is a broad-spectrum antibiotic found within μg/L concentration range in the aquatic environment. It is a known contributor of umuC induction in hospital wastewater samples. CIP can undergo photolysis to result in many transformation products (TPs) of mostly unknown toxicity. The aims of this study were to determine the genotoxicity of the UV mixtures and to understand the possible genotoxic role of the stable TPs. As such, CIP and its UV-irradiated mixtures were investigated in a battery of genotoxicity and cytotoxicity in vitro assays. The combination index (CI) analysis of residual CIP in the irradiated mixtures was performed for the umu assay. Further, Quantitative Structure-Activity Relationships (QSARs) predicted selected genotoxicity endpoints of the identified TPs. CIP achieved primary elimination after 128 min of irradiation but was not completely mineralized. Nine photo-TPs were identified. The irradiated mixtures were neither mutagenic in the Ames test nor genotoxic in the in vitro micronucleus (MN) test. Like CIP, the irradiated mixtures were umuC inducing. The CI analysis revealed that the irradiated mixtures and the corresponding CIP concentration in the mixtures shared similar umuC potentials. QSAR predictions suggested that the TPs may be capable of inducing chromosome aberration, MN in vivo, bacterial mutation and mammalian mutation. However, the experimental testing for a few genotoxic endpoints did not show significant genotoxic activity for the TPs present as a component of the whole mixture analysis and therefore, further genotoxic endpoints may need to be investigated to fully confirm this.

Taking Advantage of Databases

Toxicity databases are a useful resource to support hazard and risk assessment. They are used to retrieve historical studies for compounds of interest and to support toxicity predictions where no data exists. Toxicity predictions are either based upon study results from similar chemicals or prediction models built from these databases.

Large-Scale Predictive Drug Safety: From Structural Alerts to Biological Mechanisms

The recent explosion of data linking drugs, proteins, and pathways with safety events has promoted the development of integrative systems approaches to large-scale predictive drug safety. The added value of such approaches is that, beyond the traditional identification of potentially labile chemical fragments for selected toxicity end points, they have the potential to provide mechanistic insights for a much larger and diverse set of safety events in a statistically sound nonsupervised manner, based on the similarity to drug classes, the interaction with secondary targets, and the interference with biological pathways. The combined identification of chemical and biological hazards enhances our ability to assess the safety risk of bioactive small molecules with higher confidence than that using structural alerts only. We are still a very long way from reliably predicting drug safety, but advances toward gaining a better understanding of the mechanisms leading to adverse outcomes represent a step forward in this direction.

Computational methods for fragment-based ligand design: growing and linking

Fragment-based drug design has proved itself as a powerful technique for increasing the sampling and diversity of chemical space and enabling the design of novel leads and compounds. Computational techniques for identifying fragments, binding sites and particularly for linking, growing, and evolving fragments play a significant role in the process. Information from ADME studies and clustering property information in the form of toxicophores and chemotypes can play a significant role in aiding the design of novel, selective fragments with good activity profiles.

Monitoring of methotrexate chlorination in water

Anti-cancer drugs are an important class of pharmaceutical products. Methotrexate (MTX) is a folic acid antagonist used in high doses as antimetabolite in anti-cancer treatment as well as in low doses for the treatment of rheumatoid arthritis and adults' psoriasis. In the past, several anti-cancer drugs, including methotrexate, have been found in the environment. Their presence in water, especially if used for the production of drinking water, is even in low concentrations of particular interest, due to the risk to retrieve them in the consumed water and their high activity and grave effects. But prior to usage as drinking water, raw waters are treated and chlorination is a common practice in several countries. As such a treatment can lead to the formation of organochlorine in water, the study of the fate of MTX during chlorination in a batch trial was carried out. The reaction was monitored by dissolved organic carbon (DOC) and by fluorescence and UV spectroscopy. Investigation of by-products formed was done with liquid chromatography/mass spectrometry (LC/MS). Under the given experimental conditions, Methotrexate was eliminated rapidly (t1/2 around 21 min). However, DOC elimination was incomplete. Monitoring with LC-MS showed the formation of a monochlorinated transformation product of MTX. In silico analysis of the proposed transformation products for different carcinogenic, mutagenic and genotoxic endpoints with different software platforms provided no clear evidence that the possible transformation products after chlorination might be more toxic than the parent compound. However, since a number of alerts is altered after chlorination, it cannot be excluded that the toxicity of these transformation products might be modulated compared with the parent compound.

Assessment of hydroxylated metabolites of polychlorinated biphenyls as potential xenoestrogens: a QSAR comparative analysis∗

Alternative methods, including quantitative structure-activity relationships (QSAR), are being used increasingly when appropriate data for toxicity evaluation of chemicals are not available. Approximately 40 mono-hydroxylated polychlorinated biphenyls (OH-PCBs) have been identified in humans. They represent a health and environmental concern because some of them have been shown to have agonist or antagonist interactions with human hormone receptors. This could lead to modulation of steroid hormone receptor pathways and endocrine system disruption. We performed QSAR analyses using available estrogenic activity (human estrogen receptor ER alpha) data for 71 OH-PCBs. The modelling was performed using multiple molecular descriptors including electronic, molecular, constitutional, topological, and geometrical endpoints. Multiple linear regressions and recursive partitioning were used to best fit descriptors. The results show that the position of the hydroxyl substitution, polarizability, and meta adjacent un-substituted carbon pairs at the phenolic ring contribute towards greater estrogenic activity for these chemicals. These comparative QSAR models may be used for predictive toxicity, and identification of health consequences of PCB metabolites that lack empirical data. Such information will help prioritize such molecules for additional testing, guide future basic laboratory research studies, and help the health/risk assessment community understand the complex nature of chemical mixtures.

Diverse small molecule inhibitors of human apurinic/apyrimidinic endonuclease APE1 identified from a screen of a large public collection

The major human apurinic/apyrimidinic endonuclease APE1 plays a pivotal role in the repair of base damage via participation in the DNA base excision repair (BER) pathway. Increased activity of APE1, often observed in tumor cells, is thought to contribute to resistance to various anticancer drugs, whereas down-regulation of APE1 sensitizes cells to DNA damaging agents. Thus, inhibiting APE1 repair endonuclease function in cancer cells is considered a promising strategy to overcome therapeutic agent resistance. Despite ongoing efforts, inhibitors of APE1 with adequate drug-like properties have yet to be discovered. Using a kinetic fluorescence assay, we conducted a fully-automated high-throughput screen (HTS) of the NIH Molecular Libraries Small Molecule Repository (MLSMR), as well as additional public collections, with each compound tested as a 7-concentration series in a 4 µL reaction volume. Actives identified from the screen were subjected to a panel of confirmatory and counterscreen tests. Several active molecules were identified that inhibited APE1 in two independent assay formats and exhibited potentiation of the genotoxic effect of methyl methanesulfonate with a concomitant increase in AP sites, a hallmark of intracellular APE1 inhibition; a number of these chemotypes could be good starting points for further medicinal chemistry optimization. To our knowledge, this represents the largest-scale HTS to identify inhibitors of APE1, and provides a key first step in the development of novel agents targeting BER for cancer treatment.

3,5-dibenzyloxy-4'-hydroxystilbene induces early caspase-9 activation during apoptosis in human K562 chronic myelogenous leukemia cells

A series of 22 stilbene derivatives based on resveratrol were synthesized incorporating acetoxy-, benzyloxy-, carboxy-, chloro-, hydroxy- and methoxy functional groups. We examined the cytotoxicity of these 22 stilbenes in human K562 chronic myelogenous leukemia cells. Only four compounds were cytotoxic namely 4'-hydroxy-3-methoxystilbene (15), 3'-acetoxy-4-chlorostilbene (19), 4'-hydroxy-3,5-dimethoxystilbene or pterostilbene (3) and 3,5-dibenzyloxy-4'-hydroxystilbene (28) with IC(50)s of 78 µM, 38 µM, 67 µM and 19.5 µM respectively. Further apoptosis assessment on the most potent compound, 28, confirmed that the cells underwent apoptosis based on phosphatidylserine externalization and loss of mitochondrial membrane potential. Importantly, we observed a concentration-dependent activation of caspase-9 as early as 2 hr with resultant caspase-3 cleavage in 28-induced apoptosis. Additionally, a structure-activity relationship (SAR) study proposed a possible mechanism of action for compound 28. Taken together, our data suggests that the pro-apoptotic effects of 28 involve the intrinsic mitochondrial pathway characterized by an early activation of caspase-9.

Identification of cytochrome P450 2D6 and 2C9 substrates and inhibitors by QSAR analysis

This paper presents four new QSAR models for CYP2C9 and CYP2D6 substrate recognition and inhibitor identification based on human clinical data. The models were used to screen a large data set of environmental chemicals for CYP activity, and to analyze the frequency of CYP activity among these compounds. A large fraction of these chemicals were found to be CYP active, and thus potentially capable of affecting human physiology. 20% of the compounds within applicability domain of the models were predicted to be CYP2C9 substrates, and 17% to be inhibitors. The corresponding numbers for CYP2D6 were 9% and 21%. Where the majority of CYP2C9 active compounds were predicted to be both a substrate and an inhibitor at the same time, the CYP2D6 active compounds were primarily predicted to be only inhibitors. It was demonstrated that the models could identify compound classes with a high occurrence of specific CYP activity. An overrepresentation was seen for poly-aromatic hydrocarbons (group of procarcinogens) among CYP2C9 active and mutagenic compounds compared to CYP2C9 inactive and mutagenic compounds. The mutagenicity was predicted with a QSAR model based on Ames in vitro test data.

Rule-Based Classification of Chemical Structures by Scaffold

Databases for small organic chemical molecules usually contain millions of structures. The screening decks of pharmaceutical companies contain more than a million of structures. Nevertheless chemical substructure searching in these databases can be performed interactively in seconds. Because of this nobody has really missed structural classification of these databases for the purpose of finding data for individual chemical substructures. However, a full deck high-throughput screen produces also activity data for more than a million of substances. How can this amount of data be analyzed? Which are the active scaffolds identified by an assays? To answer such questions systematic classifications of molecules by scaffolds are needed. In this review it is described how molecules can be hierarchically classified by their scaffolds. It is explained how such classifications can be used to identify active scaffolds in an HTS data set. Once active classes are identified, they need to be visualized in the context of related scaffolds in order to understand SAR. Consequently such visualizations are another topic of this review. In addition scaffold based diversity measures are discussed and an outlook is given about the potential impact of structural classifications on a chemically aware semantic web.

Design of a multi-purpose fragment screening library using molecular complexity and orthogonal diversity metrics

Fragment Based Drug Discovery (FBDD) continues to advance as an efficient and alternative screening paradigm for the identification and optimization of novel chemical matter. To enable FBDD across a wide range of pharmaceutical targets, a fragment screening library is required to be chemically diverse and synthetically expandable to enable critical decision making for chemical follow-up and assessing new target druggability. In this manuscript, the Pfizer fragment library design strategy which utilized multiple and orthogonal metrics to incorporate structure, pharmacophore and pharmacological space diversity is described. Appropriate measures of molecular complexity were also employed to maximize the probability of detection of fragment hits using a variety of biophysical and biochemical screening methods. In addition, structural integrity, purity, solubility, fragment and analog availability as well as cost were important considerations in the selection process. Preliminary analysis of primary screening results for 13 targets using NMR Saturation Transfer Difference (STD) indicates the identification of uM-mM hits and the uniqueness of hits at weak binding affinities for these targets.