A Structure-Activity Relationship (SAR) Analysis for the Identification of Environmental Estrogens The Categorical-SAR (cat-SAR) Approach
James Devillers · 2009
The number of environmental chemicals found to have some level of endocrine activity has led to concern about the possible adverse effects these compounds may have on human and environmental health. We describe herein the development of a new set of structure-activity relationship (SAR) models for estrogenic activity using the categorical-SAR (cat-SAR) expert system. The cat-SAR models are built through a comparison of structural features found among categorized compounds in the model’s learning set (for example, estrogenic and not estrogenic compounds). Several cat-SAR models were developed from in vitro data for 122 compounds tested for estrogenicity in the E-SCREEN assay. By leaveone-out validation, of the models for the endpoint of relative proliferative potency, the concordance between experimental and predicted results ranged from 63% to 85%, and of the models based on the endpoint of relative proliferative effect, the concordance ranged from 82% to 90%. Evaluation of two-dimensional fragments derived during the modeling process suggested the models are mechanistically sound. The models also compared similarly to previous E-SCREEN models developed by using the MCASE methodology. Based on the results described herein, the cat-SAR method would be a useful approach in screening compounds for estrogen activity as well as for investigating their mechanism of action.