Sister chromatid exchange data fit with a mixture of Poisson distributions

Bowman et al. [K.O. Bowman, Wesley Eddings, Marvin A. Kastenbaum, L. R. Shenton. Sister chromatid exchange data and Gram-Charlier series, Mutat. Res., 403 (1998) 159-169.] have shown that a Gram-Charlier modification of a negative binomial distribution gives a reasonable fit to counts of sister chromatid exchange (SCE) data originally presented by Bender et al. [M.A. Bender, R.J. Preston, R. C. Leonard, B.E. Pyatt, P.C. Gooch, On the distribution of spontaneous SCE in human peripheral blood lymphocytes, Mutat. Res., 281 (1992) 227-232. ]. Here we show that a mixture of a generalized Poisson distributions also fits the data. Advantages of the generalized Poisson mixture include a simplified model involving only four parameters which fits the data more closely according to the chi-squared goodness-of-fit criterion.