Talk:Bio::Chado::Schema

Boundaries between BCS and Modware

Where is the boundary between this and modware then? I thought the point of modware was to provide that level of compatibility. Scott 17:55, 14 April 2011 (UTC)

---

Well, Modware's not my project, but it just has a lot more stuff in it, and it uses BCS as a data access layer. It's a comprehensive, heavy data model, with high-level abstractions like 'Gene', while BCS is basically just a thin wrapper around the Chado tables. It's also Moosey, with everything that goes along with that.

Being a thin wrapper doesn't mean BCS can't have some BioPerl compatibility though. The only example so far is that BCS::Result::Sequence::Feature implements Bio::PrimarySeqI, but that's illustrative of how a thin wrapper can be Bioperl compatible. Feature's implementation of SeqI is mostly just aliasing the DBIC column accessors to the Bio::SeqI accessors: `seq()` is an alias for `residues()`, `display_id()` is an alias for `name()`, etc.

I don't think anything much more involved than that belongs in BCS though. --RobertBuels 18:15, 14 April 2011 (UTC)