Kamel Mansouri

Comparison of different approaches to define the applicability domain of QSAR models.

One of the OECD principles for model validation requires defining the Applicability Domain (AD) for the QSAR models. This is important since the reliable predictions are generally limited to query chemicals structurally similar to the training compounds used to build the model. Therefore, characterization of interpolation space is significant in defining the AD and in this study some existing descriptor-based approaches performing this task are discussed and compared by implementing them on existing validated datasets from the literature. Algorithms adopted by different approaches allow defining the interpolation space in several ways, while defined thresholds contribute significantly to the extrapolations. For each dataset and approach implemented for this study, the comparison analysis was carried out by considering the model statistics and relative position of test set with respect to the training space.

Quantitative Structure–Activity Relationship Models for Ready Biodegradability of Chemicals

The European REACH regulation requires information on ready biodegradation, which is a screening test to assess the biodegradability of chemicals. At the same time REACH encourages the use of alternatives to animal testing which includes predictions from quantitative structure-activity relationship (QSAR) models. The aim of this study was to build QSAR models to predict ready biodegradation of chemicals by using different modeling methods and types of molecular descriptors. Particular attention was given to data screening and validation procedures in order to build predictive models. Experimental values of 1055 chemicals were collected from the webpage of the National Institute of Technology and Evaluation of Japan (NITE): 837 and 218 molecules were used for calibration and testing purposes, respectively. In addition, models were further evaluated using an external validation set consisting of 670 molecules. Classification models were produced in order to discriminate biodegradable and nonbiodegradable chemicals by means of different mathematical methods: k nearest neighbors, partial least squares discriminant analysis, and support vector machines, as well as their consensus models. The proposed models and the derived consensus analysis demonstrated good classification performances with respect to already published QSAR models on biodegradation. Relationships between the molecular descriptors selected in each QSAR model and biodegradability were evaluated.

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

Predicting hepatotoxicity using ToxCast in vitro bioactivity and chemical structure.

The U.S. Tox21 and EPA ToxCast program screen thousands of environmental chemicals for bioactivity using hundreds of high-throughput in vitro assays to build predictive models of toxicity. We represented chemicals based on bioactivity and chemical structure descriptors, then used supervised machine learning to predict in vivo hepatotoxic effects. A set of 677 chemicals was represented by 711 in vitro bioactivity descriptors (from ToxCast assays), 4,376 chemical structure descriptors (from QikProp, OpenBabel, PaDEL, and PubChem), and three hepatotoxicity categories (from animal studies). Hepatotoxicants were defined by rat liver histopathology observed after chronic chemical testing and grouped into hypertrophy (161), injury (101) and proliferative lesions (99). Classifiers were built using six machine learning algorithms: linear discriminant analysis (LDA), Naïve Bayes (NB), support vector machines (SVM), classification and regression trees (CART), k-nearest neighbors (KNN), and an ensemble of these classifiers (ENSMB). Classifiers of hepatotoxicity were built using chemical structure descriptors, ToxCast bioactivity descriptors, and hybrid descriptors. Predictive performance was evaluated using 10-fold cross-validation testing and in-loop, filter-based, feature subset selection. Hybrid classifiers had the best balanced accuracy for predicting hypertrophy (0.84 ± 0.08), injury (0.80 ± 0.09), and proliferative lesions (0.80 ± 0.10). Though chemical and bioactivity classifiers had a similar balanced accuracy, the former were more sensitive, and the latter were more specific. CART, ENSMB, and SVM classifiers performed the best, and nuclear receptor activation and mitochondrial functions were frequently found in highly predictive classifiers of hepatotoxicity. ToxCast and ToxRefDB provide the largest and richest publicly available data sets for mining linkages between the in vitro bioactivity of environmental chemicals and their adverse histopathological outcomes. Our findings demonstrate the utility of high-throughput assays for characterizing rodent hepatotoxicants, the benefit of using hybrid representations that integrate bioactivity and chemical structure, and the need for objective evaluation of classification performance.

Assessing bioaccumulation of polybrominated diphenyl ethers for aquatic species by QSAR modeling

Polybrominated diphenyl ethers (PBDEs) are used as flame retardants in textiles, foams and plastics. Highly bioaccumulative with toxic effects including developmental neurotoxicity estrogen and thyroid hormones disruption, they are considered as persistent organic pollutants (POPs) and have been found in human tissues, wildlife and biota worldwide. But only some of them are banned from EU market. For the environmental fate studies of these compounds the bioconcentration factor (BCF) is one of the most important endpoints to start with. We applied quantitative structure-activity relationships techniques to overcome the limited experimental data and avoid more animal testing. The aim of this work was to assess the bioaccumulation of PBDEs by means of QSAR. First, a BCF dataset of specifically conducted experiments was modeled. Then the study was extended by predicting the bioaccumulation and biomagnification factors using some experimental values from the literature. Molecular descriptors were calculated using DRAGON 6. The most relevant ones were selected and resulting models were compared paying attention to the applicability domain.

Docking-based classification models for exploratory toxicology studies on high-quality estrogenic experimental data

BACKGROUND The ethical and practical limitation of animal testing has recently promoted computational methods for the fast screening of huge collections of chemicals. RESULTS The authors derived 24 reliable docking-based classification models able to predict the estrogenic potential of a large collection of chemicals provided by the US Environmental Protection Agency. Model performances were challenged by considering AUC, EF1% (EFmax = 7.1), -LR (at sensitivity = 0.75); +LR (at sensitivity = 0.25) and 37 reference compounds comprised within the training set. Moreover, external predictions were made successfully on ten representative known estrogenic chemicals and on a set consisting of >32,000 chemicals. CONCLUSION The authors demonstrate that structure-based methods, widely applied to drug discovery programs, can be fairly adapted to exploratory toxicology studies.

In Silico Prediction of Physicochemical Properties of Environmental Chemicals Using Molecular Fingerprints and Machine Learning

There are little available toxicity data on the vast majority of chemicals in commerce. High-throughput screening (HTS) studies, such as those being carried out by the U.S. Environmental Protection Agency (EPA) ToxCast program in partnership with the federal Tox21 research program, can generate biological data to inform models for predicting potential toxicity. However, physicochemical properties are also needed to model environmental fate and transport, as well as exposure potential. The purpose of the present study was to generate an open-source quantitative structure-property relationship (QSPR) workflow to predict a variety of physicochemical properties that would have cross-platform compatibility to integrate into existing cheminformatics workflows. In this effort, decades-old experimental property data sets available within the EPA EPI Suite were reanalyzed using modern cheminformatics workflows to develop updated QSPR models capable of supplying computationally efficient, open, and transparent HTS property predictions in support of environmental modeling efforts. Models were built using updated EPI Suite data sets for the prediction of six physicochemical properties: octanol-water partition coefficient (logP), water solubility (logS), boiling point (BP), melting point (MP), vapor pressure (logVP), and bioconcentration factor (logBCF). The coefficient of determination (R2) between the estimated values and experimental data for the six predicted properties ranged from 0.826 (MP) to 0.965 (BP), with model performance for five of the six properties exceeding those from the original EPI Suite models. The newly derived models can be employed for rapid estimation of physicochemical properties within an open-source HTS workflow to inform fate and toxicity prediction models of environmental chemicals.

Prediction of Estrogenic Bioactivity of Environmental Chemical Metabolites

The US Environmental Protection Agencys (EPA) Endocrine Disruptor Screening Program (EDSP) is using in vitro data generated from ToxCast/Tox21 high-throughput screening assays to assess the endocrine activity of environmental chemicals. Considering that in vitro assays may have limited metabolic capacity, inactive chemicals that are biotransformed into metabolites with endocrine bioactivity may be missed for further screening and testing. Therefore, there is a value in developing novel approaches to account for metabolism and endocrine activity of both parent chemicals and their associated metabolites. We used commercially available software to predict metabolites of 50 parent compounds, out of which 38 chemicals are known to have estrogenic metabolites, and 12 compounds and their metabolites are negative for estrogenic activity. Three ER QSAR models were used to determine potential estrogen bioactivity of the parent compounds and predicted metabolites, the outputs of the models were averaged, and the chemicals were then ranked based on the total estrogenicity of the parent chemical and metabolites. The metabolite prediction software correctly identified known estrogenic metabolites for 26 out of 27 parent chemicals with associated metabolite data, and 39 out of 46 estrogenic metabolites were predicted as potential biotransformation products derived from the parent chemical. The QSAR models estimated stronger estrogenic activity for the majority of the known estrogenic metabolites compared to their parent chemicals. Finally, the three models identified a similar set of parent compounds as top ranked chemicals based on the estrogenicity of putative metabolites. This proposed in silico approach is an inexpensive and rapid strategy for the detection of chemicals with estrogenic metabolites and may reduce potential false negative results from in vitro assays.

A comparison of three liquid chromatography (LC) retention time prediction models

High-resolution mass spectrometry (HRMS) data has revolutionized the identification of environmental contaminants through non-targeted analysis (NTA). However, chemical identification remains challenging due to the vast number of unknown molecular features typically observed in environmental samples. Advanced data processing techniques are required to improve chemical identification workflows. The ideal workflow brings together a variety of data and tools to increase the certainty of identification. One such tool is chromatographic retention time (RT) prediction, which can be used to reduce the number of possible suspect chemicals within an observed RT window. This paper compares the relative predictive ability and applicability to NTA workflows of three RT prediction models: (1) a logP (octanol-water partition coefficient)-based model using EPI Suite™ logP predictions; (2) a commercially available ACD/ChromGenius model; and, (3) a newly developed Quantitative Structure Retention Relationship model called OPERA-RT. Models were developed using the same training set of 78 compounds with experimental RT data and evaluated for external predictivity on an identical test set of 19 compounds. Both the ACD/ChromGenius and OPERA-RT models outperformed the EPI Suite™ logP-based RT model (R2 = 0.81-0.92, 0.86-0.83, 0.66-0.69 for training-test sets, respectively). Further, both OPERA-RT and ACD/ChromGenius predicted 95% of RTs within a ± 15% chromatographic time window of experimental RTs. Based on these results, we simulated an NTA workflow with a ten-fold larger list of candidate structures generated for formulae of the known test set chemicals using the U.S. EPAs CompTox Chemistry Dashboard (https://comptox.epa.gov/dashboard), RTs for all candidates were predicted using both ACD/ChromGenius and OPERA-RT, and RT screening windows were assessed for their ability to filter out unlikely candidate chemicals and enhance potential identification. Compared to ACD/ChromGenius, OPERA-RT screened out a greater percentage of candidate structures within a 3-min RT window (60% vs. 40%) but retained fewer of the known chemicals (42% vs. 83%). By several metrics, the OPERA-RT model, generated as a proof-of-concept using a limited set of open source data, performed as well as the commercial tool ACD/ChromGenius when constrained to the same small training and test sets. As the availability of RT data increases, we expect the OPERA-RT models predictive ability will increase.

A systematic evaluation of analogs and automated read-across prediction of estrogenicity: A case study using hindered phenols

Read-across is an important data gap filling technique used within category and analog approaches for regulatory hazard identification and risk assessment. Although much technical guidance is available that describes how to develop category/analog approaches, practical principles to evaluate and substantiate analog validity (suitability) are still lacking. This case study uses hindered phenols as an example chemical class to determine: (1) the capability of three structure fingerprint/descriptor methods (PubChem, ToxPrints and MoSS MCSS) to identify analogs for read-across to predict Estrogen Receptor (ER) binding activity and, (2) the utility of data confidence measures, physicochemical properties, and chemical R-group properties as filters to improve ER binding predictions. The training dataset comprised 462 hindered phenols and 257 non- hindered phenols. For each chemical of interest (target), source analogs were identified from two datasets (hindered and non-hindered phenols) that had been characterized by a fingerprint/descriptor method and by two cut-offs: (1) minimum similarity distance (range: 0.1 - 0.9) and, (2) N closest analogs (range: 1 - 10). Analogs were then filtered using: (1) physicochemical properties of the phenol (termed global filtering) and, (2) physicochemical properties of the R-groups neighboring the active hydroxyl group (termed local filtering). A read-across prediction was made for each target chemical on the basis of a majority vote of the N closest analogs. The results demonstrate that: (1) concordance in ER activity increases with structural similarity, regardless of the structure fingerprint/descriptor method, (2) increased data confidence significantly improves read-across predictions, and (3) filtering analogs using global and local properties can help identify more suitable analogs. This case study illustrates that the quality of the underlying experimental data and use of endpoint relevant chemical descriptors to evaluate source analogs are critical to achieving robust read-across predictions.