Fig. 2

Profiling APA with Iso-Seq. a Workflow for identifying PAS sites using Iso-Seq. First, circular consensus reads were mapped to a reference genome. Next, Iso-Seq reads went through a series of two refinement steps to select for reads that contain stretches of As that are not due mispriming, resulting in a final set of reads that contain polyA tails. b Meta-gene barplots showing enrichment of signal site motifs ~20–30 nucleotides upstream of the putative cleavage site within these filtered reads. There are six consensus signal site motifs, with AATAAA being the most common. Other signal sites represented as "Other Motifs" include ATTAAA, AGTAAA, TATAAA, CATAAA, and GATAAA. AAAAAA served as a negative control. c We were able to accurately identify PAS sites and quantify PAUs for 2862 genes, which have an Iso-Seq coverage of ≥ 40 reads