Sources of variability in mutant frequency determinations in different organs of lacZ plasmid-based transgenic mice: experimental features and statistical analysis.

Sources of variability that may affect the measurement of mutant frequencies in lacZ plasmid-based transgenic mice, including rescue-to-rescue, animal-to-animal (within organ), and organ-to-organ, were studied in the context of some of the experimental features that distinguish this model from other currently used systems. Statistical analysis of repeated determinations of DNA from kidney, spleen, liver, lung, and brain indicated that 350,000 colony-forming units from each of three animals should be analyzed per treatment group in order to detect a 50% difference in mutant frequencies. Consideration is given to some of the experimental features of the transgenic assay system, including its positive selection system, its rescue efficiency, the ability to detect large deletions in the lacZ target genes, the contribution of Escherichia coli-derived mutations to the spontaneous mutant frequencies observed in vivo, and cost effectiveness of mutant frequency determinations.