PacBio Transcript Isoform Profiling

Subread filtering was performed using SMRT analysis software (v2.3.0) (Pacific Biosciences, Menlo Park, CA). Full-length read information was gathered using the RS_ReadsOfInsert.1 protocol. The reads corresponding to spliced isoforms from each device were demultiplexed, using Pacific Biosciences’ protocols and tutorials (https://github.com/PacificBiosciences/cDNA_primer/), and classified into CCS (Circular Consensus Sequences) and non-CCS subreads by ToFu^{54 (link)}. The isoform-level clustering algorithm ICE (Iterative Clustering for Error Correction) was run and the results were polished using Quiver. Transcripts lacking the 5′ end of exon 1 or the 3′ end of exon 4 were removed. Reads were mapped to the ASD to generate spliced isoform profiles and relative abundance estimates. Only transcripts accounting for 1% or more of the total number of reads are reported. The number of reads and the sequence corresponding to each spliced isoform generated by an ASD is available in the Supplementary Data 1 file.