Premium
The comparison of paired compartments by autoradiography: a methodological note
Author(s) -
Reddingius J.,
Vries O. M. H. de
Publication year - 1986
Publication title -
journal of microscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.569
H-Index - 111
eISSN - 1365-2818
pISSN - 0022-2720
DOI - 10.1111/j.1365-2818.1986.tb02786.x
Subject(s) - poisson distribution , null hypothesis , basis (linear algebra) , mathematics
summary Biochemical activities in adjacent cells, or nuclei, or other ‘compartments’, are sometimes compared on the basis of the number of silver grains visible on autoradiographs. In this report, the problem of how to test statistically the null hypothesis that the differences between the numbers of grains in two adjacent compartments are due to chance only, if one has available a sample of absolute values of such differences is discussed. The Kolmogorov test for discrete distributions is difficult to apply. More simple tests using approximations by means of the normal and chi square distributions are described. These tests are valid if, and only if, it may be assumed that the numbers of grains in the compartments are independent random variables with Poisson distributions. In autoradiographic data on RNA synthesis rates of the partner nuclei in dikaryotic cells of Schizophyllum commune , the differences appeared in general to be much smaller than expected on the basis of this assumption and the null hypothesis. This suggests that the numbers of grains in these autoradiographs follow a regular distribution rather than a Poisson one; however, it is shown that this is not necessarily true. Theoretically, the autoradiographic process might lead to regular as well as to clustered distributions. It is concluded that assuming Poisson distributions, although reasonable at first sight, may be wrong. The theoretical basis for a statistical analysis of grain counts, therefore, is insufficient.