Fig. 2: Diverse weak splice donors and acceptors are used in the PTRE-seq library.

A Location and usage frequency of splice donors in PTRE-seq reporters. Cumulative usage was computed as the sum of observed splicing fractions for each splice donor across all reporters. Nucleotide sequences are detailed in the diagram below. Note that usage of GFP (−10, −4) and 1^st module ( + 35) donor sites are not expected to trigger non-sense mediated decay, whereas 2^nd module ( + 68) sites may. B 5’ donor site strengths of PTRE-seq splice donors and annotated human splice donors. C Representative read coverage tracts illustrating alternative usage of the GFP and P donors. D Location and usage frequency of splice acceptors in PTRE-seq reporters. Cumulative usage was computed as the sum of observed splicing fractions at each corresponding splice acceptor across all reporters. Sequence details of each acceptor are shown at bottom. E 3’ acceptor site strengths of PTRE-seq splice acceptors and annotated human splice acceptors. F Synonymous mutations that abolish GFP cryptic splice donors abrogate splicing in representative reporters. Schematics of mutations are shown at left. RT-PCR analysis of recoded ASAS and AALB reporters is shown at right. BBBB reporter is included as an unspliced control. Reporters were individually transfected into HeLa cells, followed by RT-PCR and resolved on an agarose gel. Quantification shown below the gel represents the median and standard deviation over 3 biological replicates. For (B, E), splice site strength is quantified by using Maximum Entropy score⁵⁸. Uncropped gels are provided in the Source Data file.