I would like to ask if anyone has experience in running a subset of the PASA pipeline, in particular for the reconciliation of some experimental 'transcripts' with the reference annotation.

In more detail, I am working with RNA-seq data from D. melanogaster. I have reconstructed the 'transcripts' using Trinity. I have aligned these 'transcripts' to the reference genome using GMAP. Now I would like to match these 'experimental transcripts' with the reference annotation, to see how they compare.

I was wondering if I can run just a subset of the whole PASA pipeline, so basically skipping the step of alignment to the reference and using the alignment file that I generated externally.

Also, I would be glad to receive more general redirecting to any released software allowing the comparison of experimental transcripts to the reference annotation.

Cross-posted on the PASA Google Group pasapipeline-users

Cross-posted on biostars

  • 1
    $\begingroup$ Go easy on the cross-posting. You've already directly asked the authors, just be patient for their reply. $\endgroup$
    – Devon Ryan
    Jul 3, 2017 at 9:13
  • 2
    $\begingroup$ Hi Amina Echhiki, thanks for your question and welcome to Bioinformatics Stack Exchange. As you've included this (or a similar question) on other sites, it would be appreciated if you could report on the answers you have received as your own answer to this question. Stack Exchange encourages people to answer their own questions as they gain knowledge about the subject. $\endgroup$
    – gringer
    Jul 5, 2017 at 2:27

1 Answer 1


Answer from the authors:

PASA ends up using gff3 format instead of sam format for uploading the alignments. If somebody has the gmap gff3 output, then can be done to just upload that directly using the custom alignment importer, but not with sam. They suggest just having PASA rerun GMAP as part of its regular routine.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.