search

EI-CoreBioinformatics / minos

The labyrinth king judges your gene models.
GNU Lesser General Public License v3.0
9 stars 1 forks source link

Replace full-lengther classification #10

Open cschu opened 4 years ago

cschu commented 4 years ago

currently no full-lengther classification is being performed by gmc, maybe could use something else?

swarbred commented 4 years ago

Hi @cschu

So we removed full-lengther given the resources required to run, within the EI annotation pipeline we have an issue addressing this

https://github.com/EI-CoreBioinformatics/ei-annotation/issues/15

Not sure how much further luca got with this after the last comment. Potentially this could be used in the context of gmc, using the best query and target coverage as we currently do gives us a basic metric to assess transcripts against but it isn't ideal (hence why we made use of fln). It's less of an issue for the external metrics in gmc as the intention is that we generate q and t cov for each species.

lucventurini commented 4 years ago

Unfortunately I had not worked on it since, but I can definitely pick it up again now.