A STRATEGY FOR RANKING ENVIRONMENTALLY OCCURRING CHEMICALS .5. THE DEVELOPMENT OF 2 GENOTOXICITY QSARS FOR HALOGENATED ALIPHATICS

A strategy for systematic analysis and priority ranking of chemicals was applied to a class of 58 saturated halogenated aliphatics. The training set, selected by statistical design, was subjected to biological testing in an assay (the DNA precipitation assay) estimating DNA damage in V79 Chinese hamster cells. The obtained genotoxicity data were modeled by construction of two multivariate quantitative structure-activity relationships (QSARs). To experimentally validate the two derived QSAR models, a validation set of chemicals was chosen (also according to statistical design). Predictions were made for the validation set compounds, and they were found to compare well with the observed data. Thus, the two QSARs were utilized to predict the genotoxicity of 36 nontested compounds belonging to the examined class. In addition, it was found that the genotoxicity data required a multitude of chemical and structural descriptors in order to be modelable, and that log P in particular was an informative descriptor variable.