[Bioperl-l] biosql odditties part 2

[Elia Stupka]

I was thinnking why would people would store the same sequence twice, and
it seems to me the common scenario (which happens to us) is that you want
to slice databases in different ways. i.e. sometimes you want all
swissprot, sometimes you want human only swissprot,etc. and treat them as
databases. Or from the opposite perspective, you usually load sequences
from files that represent different slices.

This prompts me to the idea that it'd be neat to automagically represent
slices one is interested in within biosql. These could be completely
arbitrary (accession A,B,C) or based on an biosql-intrinsic criteria (all
entries taht come from this organism/tissue/etc)

This would allow people to always keep their uber-databases nicely up to
date, and for example, keep their "field of interest database" neatly
updated too.

I am happy to code this, let me know if you have ideas about the design of
this in terms of classes and schemas.

Elia


-- 
********************************
* http://www.fugu-sg.org/~elia *
* tel:    +65 6874 1467        *
* mobile: +65 9030 7613        *
* fax:    +65 6777 0402        *
********************************