34
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Machine Learning of Toxicological Big Data Enables Read-Across Structure Activity Relationships (RASAR) Outperforming Animal Test Reproducibility

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Earlier we created a chemical hazard database via natural language processing of dossiers submitted to the European Chemical Agency with approximately 10 000 chemicals. We identified repeat OECD guideline tests to establish reproducibility of acute oral and dermal toxicity, eye and skin irritation, mutagenicity and skin sensitization. Based on 350–700+ chemicals each, the probability that an OECD guideline animal test would output the same result in a repeat test was 78%–96% (sensitivity 50%–87%). An expanded database with more than 866 000 chemical properties/hazards was used as training data and to model health hazards and chemical properties. The constructed models automate and extend the read-across method of chemical classification. The novel models called RASARs (read-across structure activity relationship) use binary fingerprints and Jaccard distance to define chemical similarity. A large chemical similarity adjacency matrix is constructed from this similarity metric and is used to derive feature vectors for supervised learning. We show results on 9 health hazards from 2 kinds of RASARs—“Simple” and “Data Fusion”. The “Simple” RASAR seeks to duplicate the traditional read-across method, predicting hazard from chemical analogs with known hazard data. The “Data Fusion” RASAR extends this concept by creating large feature vectors from all available property data rather than only the modeled hazard. Simple RASAR models tested in cross-validation achieve 70%–80% balanced accuracies with constraints on tested compounds. Cross validation of data fusion RASARs show balanced accuracies in the 80%–95% range across 9 health hazards with no constraints on tested compounds.

          Related collections

          Most cited references40

          • Record: found
          • Abstract: not found
          • Article: not found

          Toxicology for the twenty-first century.

            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found
            Is Open Access

            Assessing skin sensitization hazard in mice and men using non-animal test methods.

            Sensitization, the prerequisite event in the development of allergic contact dermatitis, is a key parameter in both hazard and risk assessments. The pathways involved have recently been formally described in the OECD adverse outcome pathway (AOP) for skin sensitization. One single non-animal test method will not be sufficient to fully address this AOP and in many cases the use of a battery of tests will be necessary. A number of methods are now fully developed and validated. In order to facilitate acceptance of these methods by both the regulatory and scientific communities, results of the single test methods (DPRA, KeratinoSens, LuSens, h-CLAT, (m)MUSST) as well for a the simple '2 out of 3' ITS for 213 substances have been compiled and qualitatively compared to both animal and human data. The dataset was also used to define different mechanistic domains by probable protein-binding mechanisms. In general, the non-animal test methods exhibited good predictivities when compared to local lymph node assay (LLNA) data and even better predictivities when compared to human data. The '2 out of 3' prediction model achieved accuracies of 90% or 79% when compared to human or LLNA data, respectively and thereby even slightly exceeded that of the LLNA.
              Bookmark
              • Record: found
              • Abstract: not found
              • Article: not found

              Chemical regulators have overreached.

                Bookmark

                Author and article information

                Journal
                Toxicol Sci
                Toxicol. Sci
                toxsci
                Toxicological Sciences
                Oxford University Press
                1096-6080
                1096-0929
                September 2018
                11 July 2018
                11 July 2018
                : 165
                : 1
                : 198-212
                Affiliations
                [1 ]Johns Hopkins University, Bloomberg School of Public Health, Center for Alternatives to Animal Testing (CAAT), Baltimore, Maryland 21205
                [2 ]ToxTrack, Baltimore, Maryland 21209
                [3 ]UL Product Supply Chain Intelligence, Underwriters Laboratories (UL), Northbrook, Illinois 60062
                [4 ]University of Konstanz, CAAT-Europe, Konstanz 78464, Germany
                Author notes
                To whom correspondence should be addressed. Fax: +1 410 614 2871; E-mail: THartun1@ 123456jhu.edu .
                Article
                kfy152
                10.1093/toxsci/kfy152
                6135638
                30007363
                5359818e-82fe-47c9-a26f-b8f0c1ec530b
                © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Toxicology

                This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

                History
                Page count
                Pages: 15
                Funding
                Funded by: NIEHS 10.13039/100000066
                Award ID: T32 ES007141
                Funded by: EU-ToxRisk
                Funded by: European Commission 10.13039/501100000780
                Award ID: 681002
                Categories
                Machine Learning for Read-across Toxicity Testing

                Pharmacology & Pharmaceutical medicine
                Pharmacology & Pharmaceutical medicine

                Comments

                Comment on this article