[Gmod-help] polypeptide view in Gbrowse
Scott Cain
cain.cshl at gmail.com
Thu Dec 18 10:52:04 EST 2008
Hi Newton,
The cds glyph is behaving exactly as expected, and to me, represents
the best way of displaying translation information: it encodes reading
frame when you aren't zoomed in far enough to see individual residues
and knows about how to break at intron boundaries.
As for the protein glyph: honestly, before today, I didn't know it
existed (sorry Aaron :-) Looking at the code for the protein glyph,
it is assuming that the sequence it is getting and the coordinates
that it is using are amino acid residues, not base pairs. It must
have been written with the expectation of using it in the context of
GBrowse as a protein browser, which some people do. While it would be
possible to modify the protein glyph to work in a DNA context, it
would take some work. It would have to:
* detect that it has DNA instead of the expected amino acids
* do the math to figure proper placement of the letters in the glyph
(this code could likely be borrowed from other translation glyphs)
* be aware for frame/phase and intron boundaries like the cds glyph.
I'll modify the documentation of the protein glyph to make it clear
what context it is currently working in.
Scott
On Wed, Dec 17, 2008 at 3:12 PM, Newton Vidal <nwvidal at gmail.com> wrote:
> Hi, Scott and Lincoln.
>
> I have another question, related to my problem, but arising another issue:
>
> I have configured two different glyphs, one as CDS and the other as protein.
> When I zoomed in the main Gbrowse screen, the CDS glyph is shown as protein
> sequence and the protein is shown as nucleotide sequence. Is this behavior
> expected? Why the CDS, that codes for a protein, but is composed of DNA is
> shown as protein and the protein glyph is not?
>
> Representative pictures are attached.
>
>
> Thanks in advance.
>
> Newton.
>
>
> On Tue, Dec 16, 2008 at 8:52 PM, Dave Clements, GMOD Help Desk
> <gmodhelp at googlemail.com> wrote:
>>
>> Hi Newton, Scott, Lincoln,
>>
>> The draw_protein_target option of the segments glyph did not work as
>> of October. I don't recall any recent problems (or success either)
>> with the draw_target option of the transcript glyph, although it
>> wouldn't surprise me if both options use the same code. (See
>> http://thread.gmane.org/gmane.science.biology.gmod.gbrowse/808 for
>> more on draw_protein_target problems).
>>
>> I also loaded the GFF3 file for Tcruzi into a mysql database. I did
>> not get any load errors (or Apache runtime errors), but nothing
>> displays for me either. All landmarks are unrecognized. I tried
>> playing around with the conf file but to no avail.
>>
>> Dave C.
>>
>>
>> Was this helpful? Let us know at http://gmod.org/wiki/Help_Desk_Feedback
>>
>>
>>
>> On Tue, Dec 16, 2008 at 1:02 PM, Scott Cain <cain.cshl at gmail.com> wrote:
>> > Hi Lincoln and Newton,
>> >
>> > I was under the impression that the draw_protein option was
>> > incompletely implemented, but if it works with other database
>> > adaptors, perhaps there is a bug in the Chado GBrowse adaptor. Have
>> > you tried the cds glyph, which should display residues when zoomed in
>> > far enough? If draw_protein doesn't work but the cds glyph does, it
>> > indicates a Bio::Graphics problem. If neither work, it is more likely
>> > a Chado adaptor or data problem.
>> >
>> > As for the T. cruzi problem--I don't see anything in the data that
>> > looks problematic. Where there any error messages while loading or in
>> > the apache error_log (with a preemptive note: error messages about
>> > Bio::Graphics::BrowserConfig don't mean anything)?
>> >
>> > Scott
>> >
>> >
>> > On Tue, Dec 16, 2008 at 3:35 PM, Lincoln Stein <lincoln.stein at gmail.com>
>> > wrote:
>> >> Hi Newton,
>> >>
>> >> Have you tried passing the draw_protein option?
>> >>
>> >> [Polypeptide]
>> >> feature = polypeptide
>> >> glyph = transcript
>> >> draw_protein = 1
>> >>
>> >> Also check out the "protein" and "translation" glyphs.
>> >>
>> >> The T. cruzi problem sounds like a chado issue -- I will defer to Scott
>> >> and/or David on this.
>> >>
>> >> Best,
>> >>
>> >> Lincoln
>> >>
>> >> On Tue, Dec 16, 2008 at 2:01 PM, Newton Vidal <nwvidal at gmail.com>
>> >> wrote:
>> >>>
>> >>> Dear all,
>> >>>
>> >>> I am starting to work with GMOD and Chado, and I have uploaded genbank
>> >>> files from five different protozoan genomes, using bp_genbank2gff3.pl
>> >>> and
>> >>> gmod_bulk_load_gff3.pl scripts in our local chado v. 1.0 database.
>> >>>
>> >>> I have run into two major problems when visualizing data with Gbrowse
>> >>> v.
>> >>> 1.69:
>> >>>
>> >>> · The polypeptide feature shows nucleic acid sequence instead
>> >>> of
>> >>> amino acid, when clicked;
>> >>> · For four of the genomes (Leishmania sp. and Trypanosoma
>> >>> brucei)
>> >>> the others features work fine, but for T. cruzi, no data is provided
>> >>> when I
>> >>> click its features.
>> >>>
>> >>> I have edited chado.conf file to work with my database and added the
>> >>> lines
>> >>> below:
>> >>>
>> >>> [Polypeptide]
>> >>> feature = polypeptide
>> >>> glyph = transcript
>> >>> height = 10
>> >>> key = Polypeptide
>> >>> bgcolor = green
>> >>>
>> >>> I'm sending attached T. brucei and T. cruzi's GFF files and gbrowse
>> >>> conf
>> >>> file. I have a 80MB dump file and I can share it if needed.
>> >>>
>> >>> Thanks in advance,
>> >>>
>> >>> Newton Vidal
>> >>
>> >>
>> >>
>> >> --
>> >> Lincoln D. Stein
>> >>
>> >> Ontario Institute for Cancer Research
>> >> 101 College St., Suite 800
>> >> Toronto, ON, Canada M5G0A3
>> >> 416 673-8514
>> >> Assistant: Renata Musa <Renata.Musa at oicr.on.ca>
>> >>
>> >
>> >
>> >
>> > --
>> > ------------------------------------------------------------------------
>> > Scott Cain, Ph. D. scott at scottcain
>> > dot net
>> > GMOD Coordinator (http://gmod.org/) 216-392-3087
>> > Ontario Institute for Cancer Research
>> >
>> >
>
>
--
------------------------------------------------------------------------
Scott Cain, Ph. D. scott at scottcain dot net
GMOD Coordinator (http://gmod.org/) 216-392-3087
Ontario Institute for Cancer Research
More information about the Gmod-help
mailing list