Fig. 2

Library features. Four sets of libraries were prepared from zebrafish embryo lines carrying a disrupted gene. Morphologically abnormal and normal sibling embryos were collected from in-crossed lines and total RNA extracted. For each mutant line 18 to 22 samples of indexed libraries were made and sequenced by Illumina HiSeq 2500. a The number of reads and the number mapping to the Zv9 reference genome per library are shown. The total reads, the mean per library and the standard deviation are shown on the right. b For each library (dots) the proportion of reads identified as duplicates using outer mapping coordinates alone are shown on the x-axis and after accounting for the unique molecule identifier (UMI) on the y-axis. c The reads were passed through the DeTCT analysis pipeline. The number of read 2s mapped and the number of counts called under peaks as discrete regions are shown per library sample. The total reads, the mean per library and the standard deviation are shown on the right. d Using the gene list output from the DeTCT pipeline the chart shows the number of discrete regions identified per collection of libraries, the number of genes with an adjusted p-value (apv) from DESeq2 (i.e. those not removed due to low mean counts), the number of genes with an adjusted p-value <= 0.05 and the number of genes with an adjusted p-value <= 0.05 plus a fold change (FC) > 2. These analyses were performed with a stringent (-100 to +100) or relaxed (-100 to +5000) proximity filter between the Ensembl transcript and TC 3′ end