BackgroundOne of the main goals of the new chemical regulation REACH (Registration, Evaluation and Authorization of Chemicals) is to fulfill the gaps in data concerned with properties of chemicals affecting the human health. (Q)SAR models are accepted as a suitable source of information. The EU funded CAESAR project aimed to develop models for prediction of 5 endpoints for regulatory purposes. Carcinogenicity is one of the endpoints under consideration.ResultsModels for prediction of carcinogenic potency according to specific requirements of Chemical regulation were developed. The dataset of 805 non-congeneric chemicals extracted from Carcinogenic Potency Database (CPDBAS) was used. Counter Propagation Artificial Neural Network (CP ANN) algorithm was implemented. In the article two alternative models for prediction carcinogenicity are described. The first model employed eight MDL descriptors (model A) and the second one twelve Dragon descriptors (model B). CAESAR's models have been assessed according to the OECD principles for the validation of QSAR. For the model validity we used a wide series of statistical checks. Models A and B yielded accuracy of training set (644 compounds) equal to 91% and 89% correspondingly; the accuracy of the test set (161 compounds) was 73% and 69%, while the specificity was 69% and 61%, respectively. Sensitivity in both cases was equal to 75%. The accuracy of the leave 20% out cross validation for the training set of models A and B was equal to 66% and 62% respectively. To verify if the models perform correctly on new compounds the external validation was carried out. The external test set was composed of 738 compounds. We obtained accuracy of external validation equal to 61.4% and 60.0%, sensitivity 64.0% and 61.8% and specificity equal to 58.9% and 58.4% respectively for models A and B.ConclusionCarcinogenicity is a particularly important endpoint and it is expected that QSAR models will not replace the human experts opinions and conventional methods. However, we believe that combination of several methods will provide useful support to the overall evaluation of carcinogenicity. In present paper models for classification of carcinogenic compounds using MDL and Dragon descriptors were developed. Models could be used to set priorities among chemicals for further testing. The models at the CAESAR site were implemented in java and are publicly accessible.

The new European chemicals regulation Registration, Evaluation, Authorization and Restriction of Chemicals entered into force in June 2007 and accelerated the development of quantitative structure-activity relationship (QSAR) models for a variety of endpoints, including carcinogenicity. Here, we would like to present quantitative (continuous) and qualitative (categorical) models for non-congeneric chemicals for prediction of carcinogenic potency. A dataset of 805 substances was obtained after a preliminary screening of findings of rodent carcinogenicity for 1,481 chemicals accessible via Distributed Structure-Searchable Toxicity (DSSTox) Public Database Network originated from the Lois Gold Carcinogenic Potency Database (CPDB). Twenty seven two-dimensional MDL descriptors were selected using Kohonen mapping and principal component analysis. The counter propagation artificial neural network (CP ANN) technique was applied. Quantitative models were developed exploring the relationship between the experimental and predicted carcinogenic potency expressed as a tumorgenic dose TD(50) for rats. The obtained models showed low prediction power with correlation coefficient less than 0.5 for the test set. In the next step, qualitative models were developed. We found that the qualitative models exhibit good accuracy for the training set (92%). The model demonstrated good predicted performance for the test set. It was obtained accuracy (68%), sensitivity (73%), and specificity (63%). We believe that CP ANN method is a good in silico approach for modeling and predicting rodent carcinogenicity for non-congeneric chemicals and may find application for other toxicological endpoints.

The regulatory agencies should fulfil the data gap in toxicity for new chemicals including nano-sized compounds, like metal oxides nanoparticles (MeO NPs) according to the registration, evaluation, authorisation and restriction of chemicals (REACH) legislation policy. This study demonstrates the perspective capability of neural network models for prediction of cytotoxicity of MeO NPs to bacteria Escherichia coli (E. coli) for the widest range of metal oxides extracted from Periodic table. The counter propagation artificial neural network (CP ANN) models for prediction of cytotoxicity of MeO NPs for data sets of 17, 36 and 72 metal oxides were employed in the study. The cytotoxicity of studied metal oxide NPs was correlated with (i) χ-metal electronegativity (EN) by Pauling scale and composition of metal oxides characterised by (ii) number of metal atoms in oxide, (iii) number of oxygen atoms in oxide and (iv) charge of metal cation in oxide. The paper describes the models in context of five OECD principles of validation models accepted for regulatory use. The recommendations were done for the minimal number of cytotoxicity tests needs for evaluation of the large set of MeO with different oxidation states. The methodology is expected to be useful for potential hazard assessment of MeO NPs and prioritisation for further testing and risk assessment.

The aim of this article is to show the main aspects of quantitative structure activity relationship (QSAR) modeling for regulatory purposes. We try to answer the question; what makes QSAR models suitable for regulatory uses. The article focuses on directions in QSAR modeling in European Union (EU) and Russia. Difficulties in validation models have been discussed.

No abstract