Fig. 4

Sample number. The 11 mutant and 11 wild-type zmp_ph46 BAM files were downsampled and merged to create two BAM files, one from the mutant samples and one from the wild-type samples. The real index sequences in each BAM file were then replaced with fake index sequences in order to simulate assigning the reads at random to between 2 to 11 different pseudo-samples. Each of these 10 permutations was repeated 10 times, to give 100 simulations in total (shown on the x-axis), each of which was run through the DeTCT pipeline. The chart shows the average total number of counts for each pseudo-sample (diamonds on the left y-axis) and the number of (transcript counting 3′ end) TC 3′ ends with a relaxed Ensembl transcript proximity filter showing differential transcript abundance (circles on the right y-axis) with the mean above each group of ten