[Bioperl-l] Bio::Tools::Run::RemoteBlast return codes

[Scott Markel]

Chris,

It was nice meeting you and Scott C., too.  And seeing Jason again.

If you and Mark

> How about returning 1, 2, 4 for the non-zero cases, with some
> error constants set for convenience? MAJ

are okay with adding more return values, that works best for us in
Pipeline Pilot.

I'll add a Bugzilla entry.

Scott


-----Original Message-----
From: Chris Fields [mailto:cjfields at illinois.edu] 
Sent: Thursday, 14 January 2010 10:34 PM
To: Scott Markel
Cc: Bioperl-l at lists.open-bio.org
Subject: Re: [Bioperl-l] Bio::Tools::Run::RemoteBlast return codes

Scott,

I think this is fine (to change the third condition and retry with a specific code).  The other possibility is to simply throw different exceptions under each of these circumstances, which can be caught via eval to allow a retry under only certain conditions (no content, for instance).

One interesting bit: I think (though I'm not sure) the new BLAST+ allows remote BLAST queries from command line, similar to the legacy blastcl3.  Mark just wrote up a BLAST+ wrapper, so it might be worth testing that theory out.

chris

PS - BTW, nice to finally meet you at GMOD!

On Jan 14, 2010, at 4:58 PM, Scott Markel wrote:

> We've been looking at Bio::Tools::Run::RemoteBlast after some feedback
> from our customers.  Due to network irregularities (not sure what else
> to call it) users see the getting of remote BLAST results as somewhat
> random.  When results come back the hits are fine, but sometimes no
> information comes back at all.  Retrying helps.
> 
> In looking at RemoteBlast.pm there are four "return -1" cases.
> 
> * $status eq 'ERROR'      (return on line 614)
> * $line =~ /ERROR/I       (return on line 628)
> * !$got_content           (return on line 648)
> * !$response->is_success  (return on line 655)
> 
> In the case of no content we'd like to retry remote BLAST.  We're happy
> to do that part in our Pipeline Pilot Perl code wrapper for the BioPerl
> module, but we only want to retry in that case, not the other three.
> 
> What would happen if that third "return -1" changed to a different
> return value?
> 
> Scott
> 
> Scott Markel, Ph.D.
> Principal Bioinformatics Architect  email:  smarkel at accelrys.com
> Accelrys (Pipeline Pilot R&D)       mobile: +1 858 205 3653
> 10188 Telesis Court, Suite 100      voice:  +1 858 799 5603
> San Diego, CA 92121                 fax:    +1 858 799 5222
> USA                                 web:    http://www.accelrys.com
> 
> http://www.linkedin.com/in/smarkel
> Vice President, Board of Directors:
>    International Society for Computational Biology
> Chair: ISCB Publications Committee
> Associate Editor: PLoS Computational Biology
> Editorial Board: Briefings in Bioinformatics
> 
> 
> 
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l