[BioPython] [PopGen] HapMap

[Peter]

On Fri, Nov 14, 2008 at 11:49 AM, Tiago Antão <tiagoantao at gmail.com> wrote:
> Talking about hapmap, and in a different front I have some code
> available to deal with HapMap. The problem is that, in order for it to
> be useful (performance), it injects all the data in an SQL database.
> That requires a schema for persistance, but I have been "ping-ponged"
> regarding where people in Biopython say that they prefer things to be
> on BioSQL and people on BioSQL say they don't care (and, this being
> voluntary work I simply don't have the patience to fight the
> bureaucracy).

I didn't say it had to be done via BioSQL - I wanted you to check that
any schema ideas wouldn't be overlapping existing work, and felt
BioSQL was the obvious place to ask.  If these schema ideas are not a
good fit to BioSQL, then that's fine.

We do have some non-BioSQL bits of Biopython using MySQL already
(perhaps not as well looked after as they should be, Bio.GFF and
Bio.DocSQL).  The trouble with any Biopython code requiring a database
is keeping that code maintained and tested is much harder -
potentially only the original developer will actually be able to test
it.

For this particular bit of HapMap code, do you need persistence?  If
all you need is an on the fly database there may be other options
(maybe sqlite - some versions of python ship with this).

Peter