Prediction of the rodent carcinogenicity of 60 pesticides by the DEREKfW expert system

In silico model for mutagenicity (Ames test), taking into account metabolism

Bacterial reverse mutation test is one of the most common methods used to address genotoxicity. The experimental test is designed to include a step to simulate mammalian metabolism. The most common metabolic activation system is the incubation with S9 fraction prepared from the livers of rodents. Usually, in silico models addressing this endpoint are developed on the basis of an overall call disregarding the fact that the toxic effect was observed before or after metabolic activation. Here, we present a new in silico model to predict mutagenicity as measured by activity in the bacterial reverse mutation test, bearing in mind the role of S9 activation to stimulate metabolism. We applied the software SARpy, which identifies structural alerts associated with the effect. Different rules codified by these structural alerts were found in case of positive or negative mutagenicity, observed in the presence or absence of the S9 fraction. These rules can be used to understand the role of metabolism in mutagenicity better. We also identified a possible association of the results from these models with carcinogenicity.

In Silico Estimation of Chemical Carcinogenicity with Binary and Ternary Classification Methods

Carcinogenicity is one of the most concerned properties of chemicals to human health, thus it is important to identify chemical carcinogenicity as early as possible. In this study, 829 diverse compounds with rat carcinogenicity were collected from Carcinogenic Potency Database (CPDB). Using six types of fingerprints to represent the molecules, 30 binary and ternary classification models were generated to predict chemical carcinogenicity by five machine learning methods. The models were evaluated by an external validation set containing 87 chemicals from ISSCAN database. The best binary model was developed by MACCS keys and kNN algorithm with predictive accuracy at 83.91 %, while the best ternary model was also generated by MACCS keys and kNN algorithm with overall accuracy at 80.46 %. Furthermore, the best binary and ternary classification models were used to estimate carcinogenicity of tobacco smoke components containing 2251 compounds. 981 ones were predicted as carcinogens by binary classification model, while 110 compounds were predicted as strong carcinogens and 807 ones as weak carcinogens by ternary classification model. The results indicated that our models would be helpful for prediction of chemical carcinogenicity.

Evaluation of QSAR models for the prediction of ames genotoxicity: a retrospective exercise on the chemical substances registered under the EU REACH regulation

We evaluated the performance of seven freely available quantitative structure-activity relationship models predicting Ames genotoxicity thanks to a dataset of chemicals that were registered under the EU Registration, Evaluation, Authorization and Restriction of Chemicals (REACH) regulation. The performance of the models was estimated according to Cooper's statistics and Matthew's Correlation Coefficients (MCC). The Benigni/Bossa rule base originally implemented in Toxtree and re-implemented within the Virtual models for property Evaluation of chemicals within a Global Architecture (VEGA) platform displayed the best performance (accuracy = 92%, sensitivity = 83%, specificity = 93%, MCC = 0.68) indicating that this rule base provides a reliable tool for the identification of genotoxic chemicals. Finally, we elaborated a consensus model that outperformed the accuracy of the individual models.

Role of in silico genotoxicity tools in the regulatory assessment of pharmaceutical impurities

The toxicological assessment of genotoxic impurities is important in the regulatory framework for pharmaceuticals. In this context, the application of promising computational methods (e.g. Quantitative Structure-Activity Relationships (QSARs), Structure-Activity Relationships (SARs) and/or expert systems) for the evaluation of genotoxicity is needed, especially when very limited information on impurities is available. To gain an overview of how computational methods are used internationally in the regulatory assessment of pharmaceutical impurities, the current regulatory documents were reviewed. The software recommended in the guidelines (e.g. MCASE, MC4PC, Derek for Windows) or used practically by various regulatory agencies (e.g. US Food and Drug Administration, US and Danish Environmental Protection Agencies), as well as other existing programs were analysed. Both statistically based and knowledge-based (expert system) tools were analysed. The overall conclusions on the available in silico tools for genotoxicity and carcinogenicity prediction are quite optimistic, and the regulatory application of QSAR methods is constantly growing. For regulatory purposes, it is recommended that predictions of genotoxicity/carcinogenicity should be based on a battery of models, combining high-sensitivity models (low rate of false negatives) with high-specificity ones (low rate of false positives) and in vitro assays in an integrated manner.

The Use of a Chemistry-based Profiler for Covalent DNA Binding in the Development of Chemical Categories for Read-across for Genotoxicity

An important molecular initiating event for genotoxicity is the ability of a compound to bind covalently with DNA. However, not all compounds that can undergo covalent binding mechanisms will result in genotoxicity. One approach to solving this problem, when in silico prediction techniques are being used, is to develop tools that allow chemicals to be grouped into categories based on their ability to bind covalently to DNA. For this analysis to take place, compounds need to be placed within categories where the trend in toxicity can be explained by simple descriptors, such as hydrophobicity. However, this can occur only when the compounds within a category are structurally and mechanistically similar. Chemistry-based profilers have the ability to screen compounds and highlight those with similar structures to a target compound, and are thus likely to act via a similar mechanism of action. Here, examples are reported to highlight how structure-based profilers can be used to form categories and hence fill data gaps. The importance of developing a well-defined and robust category is discussed in terms of both mechanisms of action and structural similarity.

Structural alerts for estimating the carcinogenicity of pesticides and biocides

More than 20 years ago, Ashby and Tennant showed the interest of structural alerts for the prediction of the carcinogenicity of chemicals. These structural alerts are functional groups or structural features of various sizes that are linked to the level of carcinogenicity of chemicals. Since this pioneering work it has been possible to refine the alerts over time, as more experimental results have become available and additional mechanistic insights have been gained. To date, one of the most advanced lists of structural alerts for evaluating the carcinogenic potential of chemicals is the list proposed by Benigni and Bossa and that is implemented as a rule-based system in Toxtree and in the OECD QSAR Application Toolbox. In order to gain insight into the applicability of this system to the detection of potential carcinogens we screened about 200 pesticides and biocides showing a high structural diversity. Prediction results were compared with experimental data retrieved from an extensive bibliographical review. The prediction correctness was only equal to 60.14%. Attempts were made to analyse the sources of mispredictions.

Exploiting PubChem for Virtual Screening

IMPORTANCE OF THE FIELD: PubChem is a public molecular information repository, a scientific showcase of the NIH Roadmap Initiative. The PubChem database holds over 27 million records of unique chemical structures of compounds (CID) derived from nearly 70 million substance depositions (SID), and contains more than 449,000 bioassay records with over thousands of in vitro biochemical and cell-based screening bioassays established, with targeting more than 7000 proteins and genes linking to over 1.8 million of substances. AREAS COVERED IN THIS REVIEW: This review builds on recent PubChem-related computational chemistry research reported by other authors while providing readers with an overview of the PubChem database, focusing on its increasing role in cheminformatics, virtual screening and toxicity prediction modeling. WHAT THE READER WILL GAIN: These publicly available datasets in PubChem provide great opportunities for scientists to perform cheminformatics and virtual screening research for computer-aided drug design. However, the high volume and complexity of the datasets, in particular the bioassay-associated false positives/negatives and highly imbalanced datasets in PubChem, also creates major challenges. Several approaches regarding the modeling of PubChem datasets and development of virtual screening models for bioactivity and toxicity predictions are also reviewed. TAKE HOME MESSAGE: Novel data-mining cheminformatics tools and virtual screening algorithms are being developed and used to retrieve, annotate and analyze the large-scale and highly complex PubChem biological screening data for drug design.

Expert Systems for Toxicity Prediction

Expert systems offer the facility to predict a toxicity endpoint, as well sometimes as additional relevant information, simply by inputting the chemical structure of a compound. There is now a number of expert systems available, mostly on a commercial basis although a few are free to use or download. This chapter discusses nineteen currently available expert systems, and their performances (if known). Published studies of consensus predictions with these expert systems indicate that these give better results than do individual expert systems.

A test set of compounds with Tetrahymena pyriformis toxicities has been run through the two expert systems known to predict these toxicities; the predictions were quite good, with standard errors of prediction of 0.395 and 0.433 log unit. A further test set of compounds with local lymph node assay skin sensitisation data has been run through seven expert systems, and it was found that consensus predictions were better than were those from any individual expert system.

Prediction of carcinogenicity for diverse chemicals based on substructure grouping and SVM modeling

The Carcinogenicity Reliability Database (CRDB) was constructed by collecting experimental carcinogenicity data on about 1,500 chemicals from six sources, including IARC, and NTP databases, and then by ranking their reliabilities into six unified categories. A wide variety of 911 organic chemicals were selected from the database for QSAR modeling, and 1,504 kinds of different molecular descriptors were calculated, based on their 3D molecular structures as modeled by the Dragon software. Positive (carcinogenic) and negative (non-carcinogenic) chemicals containing various substructures were counted using atom and functional group count descriptors, and the statistical significance of ratios of positives to negatives was tested for those substructures. Very few were judged to be strongly related to carcinogenicity, among substructures known to be responsible for carcinogens as revealed from biomedical studies. In order to develop QSAR models for the prediction of the carcinogenicities of a wide variety of chemicals with a satisfactory performance level, the relationship between the carcinogenicity data with improved reliability and a subset of significant descriptors selected from 1,504 Dragon descriptors was analyzed with a support vector machine (SVM) method: the classification function (SVC) for weighted data in LIBSVM program was used to classify chemicals into two carcinogenic categories (positive or negative), where weights were set depending on the reliabilities of the carcinogenicity data. The quality and stability of the models presented were tested by performing a dual cross-validation procedure. A single SVM model as the first step was developed for all the 911 chemicals using 250 selected descriptors, achieving an overall accuracy level, i.e., positive and negative correct estimate, of about 70%. In order to improve the accuracy of the final model, the 911 chemicals were classified into 20 mutually overlapping subgroups according to contained substructures, a specific SVM model was optimized for each subgroup, and the predicted carcinogenicities of the 911 chemicals were determined by the majorities of the outputs of the corresponding SVM models. The model developed on the basis of grouping of chemicals into 20 substructures predicts the carcinogenicities of a wide variety of chemicals with a satisfactory overall accuracy of approximately 80%.

Predictivity of QSAR

A range of good quality, local QSARs for mutagenicity and carcinogenicity have been assessed and challenged for their predictivity in respect to real external test sets (i.e., chemicals never considered by the authors while developing their models). The QSARs for potency (applicable only to toxic chemicals) generated predictions 30-70% correct, whereas the QSARs for discriminating between active and inactive chemicals were 70-100% correct in their external predictions: thus the latter can be used with good reliability for applicative purposes. On the other hand internal, statistical validation methods, which are often assumed to be good diagnostics for predictivity, did not correlate well with the predictivity of the QSARs when challenged in external prediction tests. Nonlocal models for noncongeneric chemicals were considered as well, pointing to the critical role of an adequate definition of the applicability domain.