Fig. 1: 10X sNuc-seq data reveals cell-specific programs of 3’ isoforms.
From: Diverse cell-specific patterns of alternative polyadenylation in Drosophila
a Two classes of 3’ mRNA isoforms generated by alternative polyadenylation (APA). Left, tandem 3’ UTR (TUTR) isoforms encode the same open reading frame, but express different 3’ UTRs. Here, a gene model incorporates three different TUTR isoforms. Right, alternative last exon (ALE) isoforms harbor different C-terminal coding potential, in addition to non-overlapping 3’ UTR contents. While ALE isoforms could be generated as a result of splicing regulation, in at least some cases, ALE isoforms are determined via APA. Here, two examples of exon layouts that can generate distal ALE switching are shown. b Exemplar gene (mei-P26) with highly alternative TUTR isoforms evidence in bulk 3’-sequencing from the indicated tissues (black tracks), for which corresponding varieties of TUTR isoforms can be found in individual cell types originating from the same tissues (red tracks). c Exemplar gene (sgg) with diverse ALE isoforms (three distinct coding potentials), of which several of these are also subject to TUTR isoform generation; these collectively generate many different 3’ UTRs. Again, individual cell types from each of these tissues recapitulate the ALE and TUTR diversity seen in bulk 3’-seq data. d, e Semi-quantitative RT-PCR validation for tissue-specific expression of different 3’ UTR isoforms of mei-P26 (d) and sgg (e). Data is representative of 2 independent experiments; amplicons are 100–200 bp. Source data are provided as a Source Data file. f Aggregate density of expression data in the vicinity of known mRNA 3’ cleavage sites, across various library types. Top three plots, Drosophila data; bottom plot; mouse data. Single cell SMART-seq2 data is globally depleted in the vicinity of cleavage sites, whereas bulk 3’-seq data peak at polyadenylation sites. Single nucleus (10X-3’ kit) data also peak in the vicinity of mRNA 3’ termini, but clusters are shifted ~200–250 nt upstream of 3’-seq data. Single cell sciRNA-seq3 mouse data peak closer to 3’ ends than does the fly 10X-3’ data.
