STATISTICAL DESIGN AND ANALYSIS OF MUTATION STUDIES IN TRANSGENIC MICE

We have been working on identifying sources of variability in data fro m transgenic mouse mutation essays in order to develop appropriate sta tistical methods and designs for routine studies. Data from our lab an d elsewhere point to the presence of significant animal-to-animal vari ability, which must be taken into account in statistical hypothesis te sts. Here, the usual Cochran-Armitage (CA) test for trend in mutant fr equencies, which takes the transgene as the experimental unit, and a g eneralized Cochran-Armitage test (GCA), which fakes the animal as the experimental unit, are contrasted in computer simulations that help to quantify the differences between these statistical tests. The simulat ions report the statistical power of each test to detect treatment gro up differences, and their type I error rates. We find in general that the GCA test performs poorly compared to the CA test when it is approp riate to take the transgene as the experimental unit, and the study al so uses a small number of animals. However, the CA test performs poorl y in smell group-size studies when the animal is the appropriate exper imental unit. Extensions of the computer simulations allow for identif ication of cost-effective experimental designs. The results emphasize that the benefits of using additional animals in these mutation stud i es con be realized without substantial increases in costs. Here we ill ustrate the methods for liver studies in our lab. These methods can be used to derive optimal experimental designs for any combination of sp ontaneous mutant frequency and animal-to-animal variability. (C) 1995 Wiley-Liss, Inc.