Power simulation using data from the present microarray study. The sample size assessment was conducted to determine the number of arrays required to detect a given fold change difference between exposed and control samples. Testing for each genotype and time point group resulted in four independent assessments for each fold-change analysis. Results were then averaged, and the number of arrays required to detect a given fold-change was plotted relative to the minimal detectable fold change. From upper to lower, the lines represent simulations in which 50, 100, 500, and 1000 genes were set as differentially expressed at the fold-change indicated on the x-axis.