[maker-devel] empty maker2zff output

[Carson Holt]

Alternatively you can play with the individual filtering thresholds in
maker2zff.  Given you have > 13000 models to train with, you probably want
to par that down to a couple of thousand.

‹Carson

From:  mun hua <mh.tan85 at gmail.com>
Date:  Monday, December 2, 2013 at 6:28 PM
To:  Carson Holt <carsonhh at gmail.com>
Cc:  "maker-devel at yandell-lab.org" <maker-devel at yandell-lab.org>
Subject:  Re: [maker-devel] empty maker2zff output

Hi Carson,

Thanks for the reply.
I used maker2zff -n which successfully generated zff and dna files. Since no
filters were applied, this means that no filtering of AED values were done
on the dataset. Would this give too many incorrect hints for training?

Also, after running maker2zff, I ran a fathom -gene-stats on the output
files:

8715 sequences
0.438058 avg GC fraction (min=0.347801 max=0.759494)
15186 genes (plus=7550 minus=7636)
1497 (0.098578) single-exon
13689 (0.901422) multi-exon
171.161285 mean exon (min=1 max=17109)
792.466980 mean intron (min=1 max=23667)

When I do a fathom -validate, it lists down a lot of errors, mostly on
misordered genes and overlapping exons. Is this expected? The organism I am
annotating is fish.




On Tue, Dec 3, 2013 at 7:15 AM, Carson Holt <carsonhh at gmail.com> wrote:
> Set protein2genome=1.  Even then you may noir get many results, but on a whole
> genome you should get a few hundred that match uniprot enough to make an
> annotation directly from protein.
> 
> Then with maker2zff use the -n flag to turn off all filtering, otherwise you
> won¹t get any results (too little evidence support).
> 
> For the example not working, it is because it has really long UTRs, and one of
> the default filters was set slightly too high for the example in 2.28 for it
> to give est2genome results, but SNAP should still produce results.  The filter
> was brought back down in 2.29 and 2.30.
> 
> ‹Carson
> 
> 
> From:  mun hua <mh.tan85 at gmail.com>
> Date:  Monday, December 2, 2013 at 12:18 AM
> To:  "maker-devel at yandell-lab.org" <maker-devel at yandell-lab.org>
> Subject:  Re: [maker-devel] empty maker2zff output
> 
> Hi,
> 
> I noticed that this empty output may be because I am only using SwissProt
> protein sequences as evidence, because I do not have any EST or transcript
> data generated for my organism.
> 
> My question is, how does one proceed to train and generate final annotation
> starting with only the assembled genome sequence and UniProt/SwissProt
> sequences?
> 
> Please advise.
> 
> Thanks very much.
> 
> 
> On Mon, Dec 2, 2013 at 11:53 AM, mun hua <mh.tan85 at gmail.com> wrote:
>> Hi
>> 
>> I'm trying to create training data for SNAP using the maker2zff script that
>> comes with the maker package version 2.28.
>> I'm having the same problem as mentioned in this post:
>> https://groups.google.com/forum/#!searchin/maker-devel/maker2zff/maker-devel/
>> _Pq65Fb761U/s826h5lSxnYJ
>> My genome.ann and genome.dna files are empty, but no errors were printed.
>> 
>> Any advice on this?
>> 
>> Thanks.
>> Mun Hua
> 
> _______________________________________________ maker-devel mailing list
> maker-devel at box290.bluehost.comhttp://box290.bluehost.com/mailman/listinfo/mak
> er-devel_yandell-lab.org



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20131203/0fbd47ba/attachment-0003.html>