[Biojava-l] taxa

[Matthew Pocock]

Tom Oinn wrote:
> A potentially interesting aside; a lot of work was (and is being) done
> on making our taxonomy more interesting. This has been done primarily by
> adding auxiliary data to nodes in the taxonomic tree, such as images,
> range maps and sounds where applicable. What would the best way be to
> link these kinds of data to your taxonomy objects? 

I'll make Taxa Annotatable - that way we can put all the fluffy but 
interesting stuff in as tag-value annotations.

> 
> btw - for an example go to www.ebi.ac.uk/newt and search for 'felis
> silvestris' (press the button with a * on it), this shows the kind of
> data that we collect for each node. I suggest that to be truly useful to
> us it would have to be possible to encapsulate those data... I'm happy
> to help write code to do so too.

Woah - That's realy great - lots of pretyness. If you have the time, it 
would be great to have your help with this. In practice, we will write a 
TaxaFactory implementation for each data-source type (Corba server, SQL 
schema a, SQL schema b) and off things go.

Thomas, what is the state of the registry? How feasable is it to have 
configuration stating how to access the default taxanomic, swissprot and 
pdb data? It would be so cool if on parsing a swissprot file the 
appropreate linking object was popped into its annotation bundle for the 
taxanomic database you configure as default (e.g. in-memory vs CORBA).

Thanks.

Matthew

> 
> Tom
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l@biojava.org
> http://biojava.org/mailman/listinfo/biojava-l
> 
>