Literature DB >> 12106006

Evaluation of a rule base for identifying contact allergens by using a regulatory database: Comparison of data on chemicals notified in the European Union with "structural alerts" used in the DEREK expert system.

Stephan Zinke1, Ingrid Gerner, Eva Schlede.   

Abstract

To assess the suitability of the use of structural alerts to identify the skin-sensitising properties of chemicals, the 40 originally published structural alerts for the prediction of skin-sensitisation properties used by the DEREK system (which now contains 59 structural alerts), have been evaluated against a database developed in the German Federal Institute for Health Protection of Consumers and Veterinary Medicine (BgVV), which contains data submitted under the procedure for notifying new chemicals within the European Union. The evaluation of the 40 structural alerts used in DEREK revealed that eight of the 40 structural alerts for the prediction of skin-sensitising potential could be used without any further refinement. Ten structural alerts may need further specifications or refinements in order avoid false-positive predictions - proposals for refinement are discussed. Not enough substances were found within the BgVV database (containing data for more than 1000 substances) to evaluate ten of the DEREK substructures; hence, for these structural alerts, a judgement on their suitability for prediction of skin-sensitisation properties in expert systems is not possible. For 12 structural alerts, no comparative result could be obtained, because these rules did not "fire" for any of the examined chemicals. As a general result of the evaluation process, the approach of using structural alerts for the prediction of skin-sensitising properties of chemicals proved to be reliable. Proposals are given for a refinement of the structural alerts for prediction of contact allergy used in the DEREK system. In addition, advice and several preconditions are given, that apply when trying to teach a computer system to use structural alerts to predict toxicological properties.

Entities:  

Mesh:

Substances:

Year:  2002        PMID: 12106006     DOI: 10.1177/026119290203000305

Source DB:  PubMed          Journal:  Altern Lab Anim        ISSN: 0261-1929            Impact factor:   1.303


  2 in total

1.  Simple rules defining the potential of compounds for transdermal delivery or toxicity.

Authors:  Beatrice M Magnusson; W John Pugh; Michael S Roberts
Journal:  Pharm Res       Date:  2004-06       Impact factor: 4.200

Review 2.  Use of QSARs in international decision-making frameworks to predict health effects of chemical substances.

Authors:  Mark T D Cronin; Joanna S Jaworska; John D Walker; Michael H I Comber; Christopher D Watts; Andrew P Worth
Journal:  Environ Health Perspect       Date:  2003-08       Impact factor: 9.031

  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.