Re-annotation of a previous annotation with "est2genome=1"

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re-annotation of a previous annotation with "est2genome=1"

dupim
Hi everyone,

I'm trying to annotate a de novo Drosophila genome in MAKER 2.31.9. Unfortunately, the program annotated few proteins (4081, while flies have ~13K). After some investigation, I think this occurred because I used the option "est2genome=1", that restrict the final annotation with my transcriptome file (with ~6K sequences).

I noted that my "all.maker.augustus_masked.proteins.fasta" output have ~12K genes.

Is it possible to correct this mistake without starting from the begin? I start a new run with "est2genome=0" and using the previous gff output in several options, but it seems like it will take forever to finish.

Also, would it be necessary some filtering/edition in the "all.gff file" when put it in the options like "est_gff" and "rm_gff"?

Thank you,
Eduardo

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Re-annotation of a previous annotation with "est2genome=1"

Carson Holt-2

> Is it possible to correct this mistake without starting from the begin? I start a new run with "est2genome=0" and using the previous gff output in several options, but it seems like it will take forever to finish.

If you run in the same directory as a previous run, it will reuse archived raw reports from blast, etc.


> Also, would it be necessary some filtering/edition in the "all.gff file" when put it in the options like "est_gff" and "rm_gff"?

You can try that, but you do lose some extra info that is in the raw alignment report and not in the GFF3. So it’s usually better to let MAKER do the alignment from fasta and only use GFF3 passthrough for datasets that you no longer have access to.

—Carson
 
_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org