Fig. 2From: Quantitative gene set analysis generalized for repeated measures, confounder adjustment, and continuous covariatesPower estimates in the presence of a random effect. Plot of estimated power vs. true fold change differences (effect size) using a significance threshold of 0.05 under a longitudinal gene set simulation (V I F=20.024) with five time points using a random effect for subject. The underestimation of the VIF under QuSAGE yields overly optimistic p-values. Q-Gen remains conservative for an adequate number of replicatesBack to article page