[Gmod-help] cosmoss.org Physcomitrella mod

[Daniel Lang]

Hi Dave,

thank you for your answer!

Dave Clements schrieb:
> Suzi Lewis confirmed that the new Apollo grant has been definitively 
> funded and that work start on 1 May.  (Now you've heard it second and 
> third hand. :-) I also urge you to contact Ed and Suzi.
I'll definitely will contact them once I have finished the proposal. We 
are very much interested in contributing to GMOD, because we have 
benefited so much from it so far.
To clarify: Of course, we will only release any of the modified code if 
they want us to. I do not want to go into competition or so. If Suzi 
Lewis and colleagues think it is a good idea to include our changes, it 
would be excellent. If not, we have no intention to release it elsewhere.


> Patrick Leamaire's group in Marseilles (Patrick 
> <lemaire at ibdml.univ-mrs.fr <mailto:lemaire at ibdml.univ-mrs.fr>>, ANISEED 
> Group <aniseed at ibdm.univ-mrs.fr <mailto:aniseed at ibdm.univ-mrs.fr>>), is 
> fairly far along in porting the NISEED platform to use Chado.  That will 
> include a web front end for viewing and curation.  NISEED supports 
> atlases for expression, phenotypes and cell fates.  Patrick is also very 
> interested in interaction networks.  I strongly urge you to contact 
> Patrick to let find out if what his group is doing will work for you.  
> I'm hoping to make the NISEED platform a part of GMOD.
I have not seen ANISEED so far. Wow, they're really far a long! Really 
impressive! It would be excellent if this would become part of the GMOD 
tools!

The project we plan is a bit different, though. Sadly, for 
Physcomitrella so far in situ is not established.
We need a ontology-driven gene- and gene-set-wise visualization for 
microarray, RNAseq, Proteomics and SAGE data.

The query-tool I had in mind is similar to that at ANISEED or a mart. 
Would be cool if the ANISEED interfaces could be used for that.

So far we have implemented a modperl/Ajax GO/Gene name/Reference 
Annotation tool (genonaut). If we can continue, we want to migrate that 
to Chado and transform it into a plugin architecture to visualize 
annotation like gene family, interaction, expression and other ontology 
annotation like PO, SO and GO.
Moss development and tissues are only in part covered by PO, so we are 
developing bryophyte-extensions of PO. Either we'll end up committing it 
to PO or release it as Moss Ontology.

Most of the interactions that are available for Physcomitrella are 
miRNA<->target. We want to model interactions as ontology terms as well.

So basically what our community needs is a generic OA tool (GO, SO, PO, 
interaction, expression...). Multiple genes should be annotatable at the 
same time. The basic idea would be a webstartable Java app that reads 
multiple OBO/OWL defs from remote sources and allows to assign terms, 
evidence codes, dbxrefs and raw values for a given set of genes. IO 
should be possible with either Chado or the gene ontology schema or a 
plain text GOA like format.

Does something like this exist? If not do you think it would be an 
interesting addition to GMOD?
> 
> GMOD does not (in my opinion) have an integrated solution for 
> interaction data.  We do have Pathway Tools 
> (http://gmod.org/wiki/Pathway_Tools) which covers this.  It has many 
> GMOD users, but it does not integrate with anything else in GMOD.  It is 
> also not open source.  There's also been talk of bringing NBrowse 
> (http://gmod.org/wiki/NBrowse) into GMOD, but we have no definitive 
> plans.  Anything that would help us in this area would help us.
Ah, I had the impression that NBrowse already was integrated... If you 
think it would be a good idea, we could e.g. work on a CHADO interface 
for Nbrowse, Cytoscape or other existing tools. Alternatively, a 
javascript/Bioperl-based web app in the tradition of gbrowse could also 
be something we could work on.

>     For our proposal to be successful, it would be crucial for us to know
>     what you think of these ideas and if you would support us. As always,
>     the deadline is close - the end of April.
> 
>     What we could offer would be 3 dedicated developers for at least 3-6
>     years.
> 
> 
> That's absolutely great.  This is one of our requirements that people 
> often don't think about.  See http://gmod.org/wiki/GMOD_Membership for 
> what is needed to become a part of GMOD.
Yes, I read it and we would be very proud to become part of GMOD during 
the course of the project! It is something we have to earn.

I know it is much to ask, especially on such a short notice and under 
the circumstance that we have not been working together yet.
cosmoss.org is at a cross-roads: it was a side project to make the data 
from the genome project available - now we either continue and build a 
MOD the community can use to improve the annotation or we personally 
have to move on and follow up other projects.
We and JGI are currently working on genome V2.0 to assemble WGS 
scaffolds into chromosomes and have a lower plant reference genome.

If we get the grant, we three (Andreas (Postdoc), Karol (PhD student) 
and myself (Postdoc)) finally could focus on the development of the 
MOD...

But the program we are submitting our proposal to is very competitive 
and I strongly believe that it is very important for the success of the 
proposal to have some sort of letter of support from you to show that 
you'd welcome our efforts.

I hope, this is not too much to ask and would be very grateful. I 
definitely will use this discussion to further improve the work packages.

All the best,
Daniel