Skip to main content
Fig. 8 | BMC Bioinformatics

Fig. 8

From: Robust differential expression analysis by learning discriminant boundary in multi-dimensional space of statistical attributes

Fig. 8

Evaluates RNA-seq differential expression analysis methods using the SEQC/MAQC-III dataset. The red bars indicate the average true FDRs (refer to the left y-axis). The horizontal dashed line across the figure marks the target FDR. The blue bars indicate the average number of the detected true DEFs (refer to the right y-axis). The 90% confidence intervals of the detected DEFs are marked except for those whose true FDRs exceed the target FDR by 10%. a target FDR < 0.01. b target FDR < 0.05. At both target FDR levels, most methods effectively controlled the FDR. DCT+R+NB is the most powerful one (see main text for detailed discussions)

Back to article page