Xenoestrogens are spread throughout the environment affecting our dailylives and may produce potential toxic effects on human health. The purpose ofthis study was to develop a mechanistically reliable model capable of identifyingxenoestrogens. Our hypothesis was that there are identifiable structuralcharacteristics among a diverse set of estrogen receptor ligands that differentiateestrogenic and nonestrogenic compounds. The model's learning set wasdeveloped by collecting compounds from the National Center for ToxicologicalResearch Estrogen Receptor Binding database (NCTRER) . Thecategorical-SAR (cat-SAR) expert system was used to build the models andperform leave-none-out, leave-one-out, leave-many-out and external validationsfor model analysis. The values of all validations were between 0.80 and 0.97.Based on several analyses of rational subsets of compounds included in theNCTRER based on potency or chemical structure, it was observed that thedeveloped SAR models predictivity varied across sets. This indicates thatvariability in the SAR models or the in vitro assay results themselves must beconsidered when applying SAR models for prediction or mechanistic analyses of estrogen receptor ligands. Fragment analysis was carried out to study themechanism of estrogen receptor binding, and various important fragments wereidentified that demonstrate potential structural characteristics important forbinding. Furthermore, this led to the discovery that the cat-SAR expert systemwas able to make a higher percentage of correct predictions on specific classes ofxenoestrogen expressing these key functional groups. In conclusion, thisestrogen receptor ligand model has good predictive performance and is based onmodel attributes that are mechanistically sound.
【 预 览 】
附件列表
Files
Size
Format
View
Structure-activity relationship model for estrogen receptor ligands.
PDF
download
878987797
028-85220240
