Large-sample pairwise comparisons among multinomial proportions with an application to analysis of mutant spectra

In studies examining patterns of mutational damaged the primary variables o f interest are expressed typically as discrete counts within defined catego ries of damage. The multinomial distribution is a common model for such dat a settings. Of interest is statistical comparison of the pattern, or spectr um, of mutation among the various mutant categories. A specific question in such instances is the identification of differences among the spectral cat egories, identified, e.g., via simultaneous confidence intervals for pairwi se differences in or ratios of the multinomial response probabilities. Here in, a. selection of possible methods for constructing Such intervals is des cribed and compared via computer simulations to determine which are most ap propriate for practical use in the evaluation of mutant spectra.