[maker-devel] How to address errors encountered in process of submitting genome

[Jason Stajich]

Quick answers for now.

A) you can feed maker gff to
Funannotate  or run it alone

B) I run the annotate step in funannotate but generally transfer only
swissprot annots as product desc. Have to manually edit to remove
systematic orf names
In product desc that NCBI will flag - e.g. YAL001W, AN1234, ARB_xx. You
have to edit the annotations.swissprot.txt file to use the product
descriptor if you want to promote these to full product descriptions in the
resulting  .tbl  file

May want to run iprscan locally or wait for it running remotely to get GO
assignments included.

C) you get .tbl and Fsa
Files from gag and these are processed by tbl2asn to get sqn file. All are
produced in the result file. All automatic.

Jason

On Wed, Jun 21, 2017 at 7:33 PM Glenna Kramer <glenna.kramer at utoronto.ca>
wrote:

> Thank you for the thought!
>
> So, to clarify do you use funannotate predict on the maker gff files,
> similar to the last example given here?
> https://github.com/nextgenusfs/funannotate/wiki. I'm completely willing
> to give it a shot...
>
> Is brings up other questions for me, though.  How do you do your
> functional annotation? Maker? I noticed that funannotate will do functional
> annotation, but currently was adding in my functional annotation using GAG
> when I was converting the maker gff to tbl.
>
> Also, from what I understand, funannotate will output a gbk from the gff.
> Do you have a particular file conversion tool to get that onto the sqn
> format that you've had success with?
>
> Thanks,
> Glenna
> ________________________________________
> From: Jason Stajich [jason.stajich at gmail.com]
> Sent: Wednesday, June 21, 2017 2:25 PM
> To: Glenna Kramer; maker-devel at yandell-lab.org
> Subject: Re: [maker-devel] How to address errors encountered in process of
> submitting genome
>
> Glenna -
>
> FWIW - I've switched to doing an EVM cleanup with funannotate after MAKER
> due to these issues with MAKER and fungal genomes I submit.
>
> Jason
>
> On Wed, Jun 21, 2017 at 8:51 AM Glenna Kramer <glenna.kramer at utoronto.ca
> <mailto:glenna.kramer at utoronto.ca>> wrote:
> Hi there,
>
> I am attempting to submit a fungal genome to NCBI and have run into quite
> a few errors running tbl2asn.  I know this isn't directly related to MAKER,
> but I'm hoping that someone here has been through this process and would be
> able to give some insight (or at least point me in the direction of another
> knowledgeable source)!
>
> Here is a general overview of the process that have used so far:
> 1. Converted MAKER GFF3 files to tbl files using GAG (ran options
> remove_introns_shorter_than 10 and fix_start_stop and added functional
> annotation as well).  This seems to work well to convert the GFF3 to tbl.
> 2. Use tbl2asn to convert the tbl to sqn file.  This also seems to work,
> but I am getting lots of errors in the .val output file, which I am unsure
> how to address.  There are...
>
>     56 ERROR:   SEQ_FEAT.BadTrailingHyphen
>   1525 ERROR:   SEQ_FEAT.InternalStop
>   1142 ERROR:   SEQ_FEAT.NoStop
>     10 ERROR:   SEQ_FEAT.PartialProblem
>   2368 ERROR:   SEQ_FEAT.StartCodon
>   2368 ERROR:   SEQ_INST.BadProteinStart
>   1525 ERROR:   SEQ_INST.StopInProtein
>   8821 WARNING: SEQ_FEAT.NotSpliceConsensusAcceptor
>   8543 WARNING: SEQ_FEAT.NotSpliceConsensusDonor
>     83 WARNING: SEQ_FEAT.PartialProblem
>      1 WARNING: SEQ_FEAT.ProteinNameEndsInBracket
>     19 WARNING: SEQ_FEAT.ShortExon
>    270 INFO:    SEQ_FEAT.RareSpliceConsensusDonor
>
> Also, just as a side note, has anyone tried the new table2asn_GFF
> converter that is up to convert GFF3 directly to sqn?  I was thinking that
> I would give that a shot hoping that it would help with some of these
> errors.  However, I was instantly met with an error as well.  "Too many
> positional arguments (1), the offending value: ends."
>
> Thank you so much in advance for any help you are able to give!
>
> Glenna
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com<mailto:maker-devel at box290.bluehost.com>
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org

-- 
Jason Stajich
jason.stajich at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20170622/db2cabff/attachment-0003.html>