[Gmod-help] cosmoss.org Physcomitrella mod

[Daniel Lang]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Scott,

thank you very much for your answer!

Scott Cain wrote:
> Yes, I remember talking to you, and you probably talked with Dave
> Clements as well.
Yes indeed. Andreas and me really enjoyed this first workshop we went to
and hope to visit and contribute to the next one...
> 
> Have you talked to the Apollo folks (Ed Lee and Suzi Lewis) about your
> work?  That would be the most sensible thing to do.  The good news
> that I just recently heard (second hand, so I hope it's true), is that
> they just received some funding for continued development on Apollo.
> In my mind, that would make now a good time to approach them with
> these ideas/contributions.
I will certainly contact them. The current implementation is at an early
stage and was initially purely aimed to find something that would work
for us. We are very careful to still be compatible if new versions of
Apollo are committed. It could be a nice addon feature, I believe.
> 
> I have one question (to start with :-) about what you described below:
> do you really have a way to have Apollo talk to a
> Bio::DB::SeqFeature::Store database?  That is interesting indeed,
Yes:-) We aren't running a Chado instance yet and are using
gbrowse/DB::SeqFeature to make our annotations available.

> since it stores serialized perl objects.  You must be deserializing
> them in perl and then having Apollo talk to that?  Or is Apollo
> talking to a web service that is using the perl adaptor?  And what
> about writing back?  (OK, that morphed into multiple questions!)
Our approach is to dump GFF3 using a perl script that is invoked by a
mouse over in gbrowse. The handler then creates a webstart jnlp, which
is used to start apollo.
Annotation in apollo is more or less unchanged, we read GFF3 and provide
 several additional context menu items to allow the users to choose
which actions are performed later on in the gbrowse tracks (Delete,
Replace, create new gene).
Finally we have a modified GFF3 save function that uses certificate
based authentication to write the selected list of changes back to the
gbrowse tracks.

Initially, we wanted to do all this in plain text GFF3 and the memory
adapter since we wanted to avoid implementing and dealing with
SeqFeature delete functions... But as it turns out the solution is just
too slow.
The latter part is not implemented yet. It also will include some
functionality to promote or vote for gene models directly in gbrowse...


I hope you'll find this useful. As I mentioned earlier, MODwise, the
Physcomitrella community is at an early stage and we could really
benefit from any support you could give us to convince our funding
agency that what we plan to to makes sense...


Thank you again!

All the best,
Daniel
- --
Dr. Daniel Lang
University of Freiburg, Plant Biotechnology
Schaenzlestr. 1, D-79104 Freiburg
fax:        +49 761 203 6945
phone:      +49 761 203 6989
homepage:   http://www.plant-biotech.net/
            http://www.cosmoss.org/
e-mail:     daniel.lang at biologie.uni-freiburg.de

#################################################################
Explore the moss genome
at the 3rd cosmoss.org Physcomitrella Genome Workshop from
September 13th-15th 2010 in Freiburg
https://www.cosmoss.org/physcome_project/wiki/Cosmoss_workshop
################################################################



-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkvIgBsACgkQmJnbCpJAG3CusQCfa8ptZnjmN3VcqqB470v59UUL
3XEAn2RbSvS/ErJW4z1lw4xZg05d7jJH
=cWYo
-----END PGP SIGNATURE-----