[Gmod-help] Re: [maker-devel] [Gmod-gbrowse] Best practices for generating alignments for GBrowse...

Fri Aug 14 17:01:18 EDT 2009

No doubt flybase has a much better system.  I just figured that in the event they're not able to make that code available,  you can use this method which is fast and easy.  If you want to do any downstream tracking of what changed, you would just have to parse the gff3 file MAKER makes since the raw alignments from the old annotations (even duplicate annotations) get put in the evidence tier, so you can always check what went where.

Carson

On 8/14/09 2:53 PM, "Scott Cain" <scott at scottcain.net> wrote:

Hi Carson,

I just have one comment about migrating annotations: while your method
should work just fine, I believe what FlyBase does is actually track
the change in coordinates due to reassembly, so they know exactly
where x(old_assembly) maps to in y(new_assembly).  They can use that
information to exactly map where features have moved to, and flag
features that span regions that changed for further examination.

An example of why this might be better than using computational
methods is in the case where a duplicated region originally left out
of the assembly is put in during a subsequent reassembly.  Features
that exist in that duplicated region will only map to one region if
you use MAKER, but if I am using information about how the coordinates
changed, I can flag the features in the region for reanalysis.

Scott

On Aug 14, 2009, at 4:44 PM, Carson Holt wrote:

> Yes maker will do all that.  If you have existing models and just
> want to add EST, mRNA, or protein alignments relative to the gene
> models, you can supply the existing models via GFF3 pass-through.
> Also if the flybase method for moving annotations onto a new
> assembly is not available, I've used MAKER to do this exact thing
> (probably without as many options as flybase though).  You just pass
> the old annotations in as fasta files on the EST tier and then set
> the predictor option to est2genome.  This will align the old models
> to the new assembly, verify and realign around correct splice sites
> and then create annotations directly from the alignment.
>
> When are the GMOD summer school wikis going to made publicly
> available?  It might be a good idea to look through the MAKER class
> wiki because I specifically set up an examples of passing through
> existing annotation sets and adding EST and protein data on top.  I
> also give examples of combing multiple sets of legacy annotations
> into a consensus set.
>
> Carson
>
>
> On 8/14/09 2:31 PM, "Dave Clements, GMOD Help Desk" <gmodhelp at googlemail.com
> > wrote:
>
> Hi Bob, Josh, and all MAKER folk,
>
> One of the sessions at the Americas summer school was on MAKER, an
> annotation pipeline for eukaryotes.  I believe that MAKER can do most,
> possibly all, of what you need, and it produces GFF3 to boot.  I've
> CC'd the MAKER list for confirmation (or denial :-).
>
> Your second question was: how do people deal with incremental updates
> of assemblies?  This came up at the just finished August 2009 GMOD
> Meeting.  Josh Goodman of FlyBase indicated that they had a set of
> programs that compared the two assemblies and programmatically moved
> annotations from the old to the new for almost all annotations.  For
> each migration there are also a small number of annotations that are
> flagged for manual curation and these are looked at by FlyBase
> curators.
>
> Josh, do you know if the FlyBase process might be made available
> outside of FlyBase?  Would it be useful outside of FlyBase?
>
> Do other communities have scripts for doing this type of migration?
>
> Hope this helps,
>
> Dave C.
>
> On Wed, Aug 5, 2009 at 2:11 PM, Bob Freeman<bob_freeman at hms.harvard.edu
> > wrote:
> > Hello all!
> > Our IT group is almost finished installing both Chado and GBrowse
> onto our
> > cluster and web hosting environments. Before I load any data, I was
> > wondering what programs people use for generating certain types of
> data. I
> > haven't been able to find any specific recommendations in any of
> the online
> > docs or tutorials. If I've overlooked them, my apologies.
> > Specifically, we'd like to load the following:
> > ESTs positioned against genomic scaffolds (same species)
> > BLAST? or a better program?
> > Gene models
> > These are already in GFF3 format from the model generating programs
> > Full-length mRNAs/clones (same species)
> > BLAST? or something better?
> > mRNAs from closely-related organisms for cross-species comparisons
> on gene
> > structure
> > ?
> > proteins from closely-related organisms for cross-species
> comparisons on
> > gene structure
> > ?
> > Thanks for the advice, and I welcome any additional suggestions.
> >
> > On a separate note, how do folks handle incremental updates of
> genome
> > assemblies? In the space of 6 months we've already moved to a 1.1
> assembly.
> > Should I double the amount of analyses available? or is there a
> way to have
> > Chado / GBrowse remap coordinates?
> > Thanks,
> > Bob
> > -----------------------------------------------------
> > Bob Freeman, Ph.D.
> > Acorn Worm Informatics, Kirschner lab
> > Dept of Systems Biology, Alpert 524
> > Harvard Medical School
> > 200 Longwood Avenue
> > Boston, MA  02115
> > 617/432.2293, vox
> > When choosing between two evils I always take the one I've never
> tried
> > before. -- Mae West, Klondike Annie
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> > Let Crystal Reports handle the reporting - Free Crystal Reports
> 2008 30-Day
> > trial. Simplify your report design, integration and deployment -
> and focus
> > on
> > what you do best, core application coding. Discover what's new with
> > Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> > _______________________________________________
> > Gmod-gbrowse mailing list
> > Gmod-gbrowse at lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
> >
> >
>
>
>
> --
> * Please keep responses on the list!
> * Was this helpful?  Let us know at http://gmod.org/wiki/Help_Desk_Feedback
>
> _______________________________________________
> maker-devel mailing list
> maker-devel at yandell-lab.org
> http://yandell-lab.org/mailman/listinfo/maker-devel_yandell-lab.org
>
>

-----------------------------------------------------------------------
Scott Cain, Ph. D. scott at scottcain dot net
GMOD Coordinator (http://gmod.org/) 216-392-3087
Ontario Institute for Cancer Research

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://brie4.cshl.edu/pipermail/gmod-help/attachments/20090814/0e7d985c/attachment.html>