Fig. 4
From: Power analysis for RNA-Seq differential expression studies
Critival values plot for both Wald test and LRT with equal dispersion parameters. Critical values were calculated at the nominal false positive error rate of 0.001 from empirical percentile of null statisitics at 5 different mean expression levels for both Wald test (solid line) and LRT (dashed line), and for the chi-square distribution with 1 degree of freedom (purple line). Both Wald test and LRT with the empirical distribution have larger critical values than both Wald test and LRT with the chi-square distribution, and the Wald test has much larger values than the LRT with the empirical distribution