DGE tool TPR/FPR performance with 2 replicates

This figure is an expansion on Figure 2 from our paper in RNA "How many biological replicates are needed in an RNA-seq experiment and which differential expression tool should you use?". The figure shows the comparison of the true positive rate (TPR) and false positive rate (FPR) performance for each of the DGE tools for extremely low-replication (n_r=2) RNA-seq data. As for Figure 2 in the paper, the TPRs & FPRs for each tool are calculated by comparing the mean number of true and false positive (TPs and FPs) calculated over 100 bootstrap iterations to the number of TPs and FPs calculated from the same tool using the full clean dataset (error-bars are 1 standard deviation). Although the TPRs and FPRs from each tool are calculated by comparing each tool against itself rather than a tool-independent ‘gold standard’ (albeit with the full clean dataset) the results are comparable across tools (see the paper for more details).