[BioRuby] [GSoC][NeXML and RDF API] RDF API

[Pjotr Prins]

Hi Anurag,

On Thu, Jul 15, 2010 at 08:27:45PM +0530, Anurag Priyam wrote:
> However, I was thinking of something like:
> 
> seq = Bio::Sequece.new
> seq.annotate do |graph|
>   graph << [self, CDAO[:foo], 'moo' ]
> end
> 
> seq.annotation.query :predicate => CDAO[:foo]
> 
> I think with this design we can maintain loose coupling between the rdf lib
> and bioruby components. I have just begun creating the classes to realize
> the specs, so the design can still be modified completely if I am in a wrong
> direction.

I think this is the idea. RDF generator should be generic and easily
used for extending existing objects. That is very good. In other
words, the Sequence class should not *know* about RDF - we should not
pollute existing classes (even further) with RDF knowledge, if we can
avoid it.

You can create a specialized RDF::Sequence, or RDF::Alignment, object,
which would add certain features to a base Sequence class (without
modifying the Sequence class itself, for sure). These classes should
be opaque for whether we are dealing with nucleotids, or amino acids
(is my opinion).

So the first thing to do is to write the Specs for such a system. The
current Specs are merely object invocations for RDF itself.

So what I would like to see is Specs that do something real. Rather
than the example.org URI, use something that is meaningful. Write
Specs for using real BioRuby classes and/or NeXML classes. 

Others, Raoul for one, have ideas too for RDF too. So the Specs will
help out with ideas.

Write more directed Specs and we will discuss them.

Pj.