This directory contains the genome coordinates of PacBio reads (ROIs) produced within the GENCODE Capture Long-Seq project, in GTF format.

Only reads with the following characteristics are reported:
- mapped uniquely in the genome
- if spliced, their entire chain of introns is canonical (GT|GC / AG)
- if unspliced, they must have a detectable polyA tail

In other words, the difference between this set and the HCGMs present in the parent directory, is that this set may contain non double-bounded reads.

All files correspond to genome assemblies hg38 and mm10.

# File naming scheme:

<species>All_Cap1_<tissue>_allCanonicalExons.gff.gz

where:
	species:
		"mm": mouse
		"hs": human

	tissue: self-explanatory. "Undeter": non-demultiplexed reads (i.e., of unknown tissue origin)

# File format:

Standard GTF, with exon records only. 
The originating PacBio read identifier is reported in both the "gene_id" and "transcript_id" attributes.