Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Gene set analysis using variance component tests

Figure 1

Power comparison using simulations with the true covariance to be compound symmetry. We set n = 20 and p = 40; covariance (cov) = 0.1 (top panel) and 0.5 (bottom panel). For each panel, the left plot is based on permutation and the right plot is based on the Satterthwaite scaled chi-square approximation. For each plot, the powers of TEGS assuming six different covariances were compared. TEGS assuming working independence (Indpt) corresponds to the global test of Goeman et al. (2004). The horizontal dotdash line indicates the size of the test (i.e., 5%).

Back to article page