[Bioperl-l] Blast stuff

[Steve Chervitz]

Ewan,

The merging does in fact refer to adding psiblast parsing functionality to
blast.pm so that psiblast.pm can be obsoleted and we can simplify the SearchIO
code base. Yes, this is a destabilizing change in the short-term, but
stabilizing in the long-term.

Jason and I have been talking about targeting this for 1.2, but I'm having
doubts because it is a fairly major undertaking, and there's time pressure to
release 1.2. So I think we'd all be more comfortable if it was done first in a
developer's release to allow for more testing.

As for work on the objects produced by the parsers, I've already committed code
that allows the generic search objects work with the writers (specifically to
allow GenericHits to undergo HSP tiling).

Perhaps the best strategy for 1.2 is to focus on testing the GenericHit HSP
tiling and beefing up the SearchIO docs. The psiblast.pm -> blast.pm merging
would then be deferred until after 1.2. Sound good? (BTW, what's the target
date for 1.2?)

Steve

--- Ewan Birney <birney@ebi.ac.uk> wrote:
> 
> Steve - what does merging psiblast and blast parsing in the searchIO mean?
> I don't want any destabilising moves on the code base as we head towards
> 1.2 (i know you too well steve....)
> 
> 
> So...
> 
>   Making psiblast.pm produce the same objects as blast.pm and so can be
> used by the same set of results writers - GOOD
> 
>   Killing psiblast.pm and merging into blast.pm - BAD - destablising.
> 
> 
> 
> make sure you don't leave it to the last minute to put changes in.
> remember the matrix of platforms-and-perl-versions is huge for testing.
> 
> 
> 
> 



=====
Steve Chervitz
sac@bioperl.org

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com