[Bioperl-l] Blast stuff

Jason Stajich jason@cgt.mc.duke.edu
Mon, 2 Dec 2002 21:45:47 -0500 (EST)


Jason Stajich
Duke University
jason at cgt.mc.duke.edu

On Mon, 2 Dec 2002, Ewan Birney wrote:

>
> bioperl-run:
>
> Ok, having looked at some of the other options, I reckon the simplest is
> the one lincoln suggested and backed up by Hilmar -
> Bio::Tools::RemoteBlast gets transfered back to core. Any one vote
> otherwise?
>
Sounds fine. Let's move it back since this doesn't require any local apps.
It was called Bio::Tools::Run::RemoteBlast - shall we keep it that way?


Can just move the pm,v file on the repository site and will get all the
logged changes still.

>
>   Elia/Catherine - what are your plans for bioperl-run - can one of you
> take charge of regression testing against the slowly solidyfying core and
> make a branch and tar file when the time comes.
>

Elia - I think we *will* want to tag, a branch may or may not be necessary
we can always create one if we find that we are changing the API and
back-fixing bugs pre-changes.


This will probably be post 1.2 but I'd like to encourage unifying all the
modules to inherit from Bio::Tools::Run::WrapperBase and support the
generic 'run' function which may accept parameters in as a hashref - ala
$standaloneblast->blastall($seq) becomes

$standaloneblast->run({'-program' => 'blastall',
                       '-blasttype' => 'blastp',
                       '-query'     => $seq,
                       '-db'        => 'nr'});

This makes pipeline construction easier I suspect?

>
>   [Hilmar - ditto for bioperl-db - are you ok to do this?]
>
>
> Bio::Tools::Blast.pm
>
>   I want to kill this - it has lots of confusing documentation; it is big;
> I think Bio::SearchIO is the way forward and is stable enough to take it
> on; I will bug fix Bio::SearchIO on spec for the 1.2 series (phew!). So...
> can we remove it and put in a warning in a BEGIN block saying - it has
> died, please try out Bio::SearchIO.
>
>
>   SteveC and Jason - what are your votes here.
>
>
>
> ...1.2 marches on...
>
>
> (BTW - tests pass and I have latest ExtUtils to try to find the Makefile
> fix...)
>
>
>
> ewan
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
>