In silico prediction of chemical toxicity on avian species
December 2014
East China University of Science and Technology, Shanghai, China
The aim of this study was to develop an in-silico prediction tool for chemical toxicity on avian species. Therefore, the toxicity of more than 663 diverse chemicals, including pesticides and industrial chemicals, on 17 avian species was accessed. Data sets were selected from the EPA Ecotox database and data for mallard duck and northern bobwhite quail were used to build models for avian toxicity prediction, while Japanese quail data was used to validate the models. All the chemicals were classified into three categories, i.e. highly toxic, slightly toxic and non-toxic, based on the toxicity classification criteria of the United States Environmental Protection Agency (EPA).
Chemical category approaches were used for model development and both, molecular descriptors and fingerprints, were calculated to represent the compounds. Afterwards, binary classification models were developed using machine learning methods. The best model had an overall accuracy of 0.851 for the prediction of toxicity on avian species. Furthermore, several representative substructures for characterizing avian toxicity were identified.
In silico prediction of chemical toxicity on avian species using chemical category approaches
Yun Tang, Philip W. Lee
Added on: 02-06-2024
[1] https://www.sciencedirect.com/science/article/abs/pii/S0045653514014003?via%3Dihub
