[Gmod-help] Re: GMOD wiki and GFF format.

[Dave Clements]

Hi Brian,

Thanks for the introduction and welcome to the GMOD project!  I am sure that
GBrowse will be able to meet your needs.  I am also sure that JBrowse can
meet your needs, but its ability to create a nice interface for selecting
only a few of many tracks is not as advanced as GBrowse's (Mitch and Ian may
correct me).

When you have questions, GFF and GBrowse questions are best sent to the
GBrowse list, and JBrowse questions to the JBrowse list.  If you aren't sure
where to ask, send the questions to the help desk and we will sort it out.
See http://gmod.org/wiki/GMOD_Mailing_Lists for links.

Have you ever worked with Allen Rodrigo?  He's now the director of NESCent
and I had a conversation with him just a month ago about showing HIV in
genome browsers.

Thanks,

Dave C.

On Wed, May 26, 2010 at 2:27 PM, Brian Foley PhD <btf at lanl.gov> wrote:

> Dear Dave,
>
> I just signed up to the GMOD wiki today.  I've used wikipedia a tiny bit,
> as
> far as having an account, editing a few pages, participating in user:talk
> and page discussions a little bit.  I am a biologist, not a computer
> scientist, but I have been doing "bioinformatics" DNA sequence analysis,
> GenBank annotation etc since 1984, so I am a longtime computer user.
>
> Since 1995 I have been at the HIV Sequence, Immunology and Vaccine Trials
> Databases at Los Alamos National Lab  http://www.hiv.lanl.gov
>
> I joined GMOD today, because we want to build a genome browser for HIV.  It
> makes most sense to use an existing browser, and even more sense to use an
> existing format for data the browser displays (so we can switch browsers
> easily, for example).
>
> The HIV genome is tiny, just 9.4 Kb.  We will have dense annotation of many
> sites that pertain to a single amino acid (drug resistance mutation, for
> example) or a few amino acids (antibody binding site, for example).  For
> many sites in the genome we will have 30 or more annotations (many
> monoclonal antibodies bind to the same site, or the site produces
> resistance
> to more than one drug).
>
> So our genome browser I more about "depth" than "width".  I don't think
> most
> users will want to see more than a few layers of depth at any one time.
> They will either want to see drug resistance data, or antibody data, or CTL
> epitope data, or a couple of those types of things.  But not all possible
> types of data at once.
>
> So, I envision a database of all types of data about each site in the
> genome, and then querying that database to pull a few of the fields and put
> them in GFF format for display in the browser.
>
> Looking at the GFF format at:
> http://gmod.org/wiki/GFF
> I see that I need to read it carefully and "play around" a bit before I
> will
> even know what types of questions I might have.
>
> Anyway, I wanted to introduce myself today, so that if I write to you in
> the
> future you may have already thought a little bit about other people with
> similar "deep" instead of "wide" data sets.  Some of the human genome SNP
> and quantitative expression data will make the human genome both deep and
> wide.  For a single human gene we will accumulate data about allele
> frequencies in various populations, expression levels in various tissues,
> etc...
>
> Sincerely,
> Brian Foley, PhD
> HIV Databases
> (505) 665-1970
> btf at lanl.gov
>
>
>
>


-- 
http://gmod.org/wiki/GMOD_News
http://gmod.org/wiki/Calendar
http://gmod.org/wiki/Help_Desk_Feedback
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://brie4.cshl.edu/pipermail/gmod-help/attachments/20100527/f9ee0546/attachment.html>