In collaboration with Airbus Safran Launchers Company (ASL), a research program was set up to have, in the future, optimized tools to predict the toxicity of High-Energy Materials (HEM). The European Union has adopted a legislation about chemicals named REACh (Registration, Evaluation, Authorization and Restriction of Chemicals) to ensure a high level of protection of human health and environment. REACh and the OECD (Organization for the Economic cooperation and Development) describes the internationally guidelines of the different toxicological mutagenicity tests like the Ames test, the micronucleus test and the chromosomal aberration test. Several in silico methods can be used to predict the toxicity of molecules such as (Quantitative) Structure Activity Relationship ((Q)SAR) and especially machine learning methods like neural networks. Moreover, the search of structural similarity among molecules is one of the most conventionally used tools in the in silico predictions. The database used in this study compiled available mutagenicity data for a lot of diversified molecules. An explicit list of HEMs, from which we know the experimental results of the different toxicological tests in the literature, has been chosen to consolidate the predictive system. On the basis of similarities, we select molecules from the database thanks to a smarter algorithm and we perform thereafter the prediction using an in silico methods. The goal of this study is to use a smarter algorithm to investigate the use of similarity indices and to obtain an efficient prediction tool for different kind of explosives. All these compounds were evaluated for evidence of potential mutagenicity. This prediction tool will be compared with softwares often used in the scientific community like ACD, PASS or Lazard. This phase allows assessing the prediction’s quality. The subject of this paper is to show that our similarity tool can be encoded and applied itself to a wide range of different similarity indices for the prediction of activity of chemical compounds.

Prediction of regulation toxicological tests applied to high energy molecules

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.