Skip to main content
Fig. 3 | Genome Biology

Fig. 3

From: ACE: a probabilistic model for characterizing gene-level essentiality in CRISPR screens

Fig. 3

Detection of differential essentiality in simulated data. A Performance in binary classification of differential essentiality for 300 genes that were “essential” (at various levels of simulated essentiality ϕ) in “test” samples and “nonessential” (ϕ=0) in “control” samples. Both samples also included 300 genes at strong (ϕG=0.99) and 300 genes at moderate (ϕG=0.5) essentiality, as well as a large collection (3150) of nonessential genes (ϕG=0). 300 genes from each of these uniformly essential sets were used as negative controls. B Fractions of differentially essential genes correctly detected by ACE (at a Bonferroni-corrected empirical p < 0.05) at various values of simulated essentiality ϕ. C Full ROC curves for each of the simulated differential essentiality levels. In all cases, three replicates of the control and test samples were simulated for each value of ϕ. Methods compared are as described in Fig. 2

Back to article page