constitutive.tsv alignment misannotation

[David McWatters]

Hello!

I have been taking a look at the constitutive.tsv outputs for one of my data sets from voila modulize with --keep-constitutive/--show-all on and it appears as though the reference exon positional annotation in these outputs is correct for - strand genes, but when the strand is + the coordinates are actually for the intron between two exons. 

Was hoping to clarify if this is something on my end or genuine miss-annotation. 

Thank you for the help! 

Best, 

David 

[San Jewell]

Hi David,

While running voila modulize and voila view on a small set of test data, I wasn't able to produce this behavior. Within the constitutive.tsv output, for both plus and minus strands, it doesn't seem like the reference_exon coordinate, nor spliced_with coordinate, are over the intronic region.

Would you be able to verify you are using the latest version of voila/majiq? If the problem persists, would you be able to take a screenshot of voila view in the problem area, along with the entry in the tsv file which is incorrect? If I still can't reproduce it, I may also request for a subset of your data to see if there's something strange with the input which causes the issue.

Thanks,

-San