Regulatory bodies are interested in using in silico methods to address animal welfare issues, reduce costs, and obtain information regarding chemicals that are difficult to purify for in vivo and in vitro tests. In silico models, such as quantitative structure−activity relationships (QSARs), can be used to predict the biological activities of chemicals from their structure. A type of QSARs is Ames/QSAR, which is based on the Ames test data and can predict the mutagenicity of a chemical. Ames/QSAR models are currently used to predict the mutagenicity of impurities in pharmaceuticals [1] and other chemicals, such as pesticides and their metabolites [2].
