Figure 2
From: Statistical implications of pooling RNA samples for microarray experiments
Approximately equivalent power curves under different pooling schemes. Power curves generated for two-sample t tests. Equal pooling assumed. Legend: n is the number of subjects per treatment group; c is the number of arrays per group. The five pooling schemes with different choices of number of subjects and number of arrays have approximately equivalent power curves when type I error rate is controlled at 0.05.