Skip to main content
Fig. 5 | Genome Biology

Fig. 5

From: DegNorm: normalization of generalized transcript degradation improves accuracy in RNA-seq analysis

Fig. 5

Differential expression analysis results. a Log2 ratio of false-positive rate of UQ/RUVr/TIN relative to DegNorm at q value = 0.05 criterion for data sets where genes have no differential expression. b–f DE results for SEQC-AB data. b Coefficient of variation (CV) vs. mean normalized read counts (in log scale). c Empirical cumulative distribution function (ECDF) of the p value from DE analysis. d Receiving operating characteristic curve (ROC) calculated from ~ 500 PCR-verified genes [34]. Genes with absolute log2 fold change value ≥ 2 were defined as true positives, and absolute log2 fold change value ≤ 0.1 was defined as true negatives. Genes with log2 fold change in between were disregarded. e ROC calculated based on ~ 17,304 PCR-verified genes published in a separate study [27] for the same SEQC AB samples. The same threshold values of log2 fold change as in c were used in defining the positives and negatives. f The area under the ROC curve (AUC) statistic as a function of the threshold value of absolute log2 fold change used to define the true positives based on the PCR data from e

Back to article page