Performing power analysis with powsimR

Statistical power analysis is used to determine the sample size needed to detect an effect of a certain size with a certain level of statistical significance. This is important because it allows researchers to ensure that their studies are adequately powered (that is, enough replicates have been sampled) to detect the effects that they are interested in. Without sufficient power, there is a higher risk of failing to reject the null hypothesis when it is false – that is, to miss important differentially expressed genes. In this recipe, we’ll use the powsimR package (which is not in Bioconductor) to perform two types of power analysis. Both of these will be performed with a small real dataset. First, we shall do power analysis with two treatments, test and control, and then with just one. With each, we shall estimate the replicates that are needed to spot differences in gene expression of a particular magnitude – if they’...