Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: Leveraging biological and statistical covariates improves the detection power in epigenome-wide association testing

Fig. 2

Performance evaluation of covariate-adaptive FDR control methods on simulated data (random signals). Performance is evaluated based on FDR control (average false discovery proportion, panel a) and power (average true positive rate, panel b). The performance is compared under combinations of different level of covariate informativeness (“Non-informative,” “Moderately informative,” and “Strongly informative,” columns), signal density (“Sparse signal,” “Medium signal,” and “Dense signal” rows), and signal strength (“Weak,” “Medium,” “Strong”). Error bar indicates the standard error across 100 repetitions, and the dashed line denotes the target FDR level (0.05)

Back to article page