How to determine spliced, unspliced, ambiguous read assignments for starsolo velocyto option

[Lioscro]

Hello,

I was wondering if there is a way to find out, given a mapped read in the BAM file generated using the starsolo velocyto option, which of the three species (spliced, unspliced, ambiguous) this read was assigned to?
For genes, I can simply look at the GX tag to determine what gene count the read was assigned to, but I am having some trouble figuring out if there is an analogous option for velocity counts, if one exists.

Thanks in advance!

[alexdobin]

Hi Joseph,

there is no such option at the moment, but it would be a good feature to have - adding it to my TODO list.

Cheers
Alex

[Lioscro]

Awesome. Looking forward to it 👍

In the meantime, could you help me understand how starsolo assigns a read to each of the velocity species (spliced, unspliced, ambiguous)?
I've been attempting to (until such a feature exists within STAR), iterate through each read and manually assign them to the above species. The following is what I've been going off of, after referencing how velocyto.py performs the procedure.

• Spliced: reads that have a GX tag and have an N in the cigar string
• Ambiguous: reads that have a GX tag and do not have an N in the cigar string
• Unspliced: reads that do not have the GX tag and overlap an intronic region of a single gene (if it overlaps an intronic region of multiple genes, the read is ignored)