Skip to main content
. 2019 Dec 20;20(Suppl 24):675. doi: 10.1186/s12859-019-3248-9

Fig. 3.

Fig. 3

Power by numbers of technical replicates. Power was averaged over 100 simulation runs for 6 different sample sizes of technical replicates where the number of biological replicates equals 3 and the expected mean number of false positives equals 5. Different methods were used for the degrees of freedom approximation: FMT-VC (solid blue) is the fully moderated t-test with the variance components method; FMT-Sat (dotted blue) is the fully moderated t-test with the Welch-Satterthwaite method; OT-VC (solid black) is the ordinary t-test with the variance components method; OT-Sat (dotted black) is the ordinary t-test with the Welch-Satterthwaite method; Limma (dotted dark green) is the Limma method with replicates’ correlation estimation; Dream-Sat (dotted red) is the Dream method with the Welch-Satterthwaite method