Methods for comparing Salmonella mutagenicity data sets using nonlinear models.

A variety of linear and nonlinear mathematical models have been proposed to characterize Salmonella mutagenicity data sets, but no systematic procedure has been suggested for comparing two or more data sets across experiments, laboratories, occasions, mutagens or treatment conditions. In this paper, a general method for data-set comparison is provided. Nonlinear regression techniques are applied to real data sets. Data-set and parameter equivalence are described in depth. Confidence-band construction for nonlinear models and other graphical techniques are presented as auxiliary tools. Key Statistical Analysis System (SAS) code programs are provided.