[Bioperl-l] Re: [Gmod-schema] Re: BioSQL or chado

[Allen Day]

> > There is a perl O/R layer for chado -- it's autogenerated Class::DBI code
> > from the CREATE TABLE statements.  There are not yet any adapters written
> > that convert the Class::DBI objects to/from bioperl objects.
> 
> Yes of course, sorry, totally spaced on this part, I've been focused on
> the components developed by the harvard/flybase folks.
> 
> Is there any plans to write the bioperl bridge anytime soon?

Not in the immediate future, as I don't really have an urgent need to
bridge at this time.  If I had time, I'd start with the cvterm/ontology
and rad(expression)/bioperl-microarray bridges.

It shouldn't be too hard to autogenerate the simple bridging where there
is a one-to-one correspondence between chado tables (or chado star
subschemas) and bioperl classes.  We just need to write a configuration
file that maps chado tables/columns to bioperl classes/attributes and hand
it, along with either a chado or bioperl object, to a generic factory
class that knows how to do the interconversion.

The more complex relations in chado (eg features graphs) are going to be
harder to bridge because chado and bioperl model these relations
differently.

-Allen