[Bioperl-l] Fwd: Changes to NCBI BLAST and E-utilities.

[Mauricio Herrera Cuadra]

Forwarding to the proper lists...

-------- Original Message --------
Subject: [O|B|F Helpdesk #889] Changes to NCBI BLAST and E-utilities.
Date: Fri, 5 Feb 2010 10:08:51 -0500
From: mcginnis via RT <sup...@helpdesk.open-bio.org>
Reply-To: sup...@helpdesk.open-bio.org
To: ch...@bioteam.net, hei...@sanbi.ac.za, hl...@gmx.net,
ja...@bioperl.org, maur...@open-bio.org

Fri Feb 05 10:08:51 2010: Request 889 was acted upon.
Transaction: Ticket created by mcgi...@ncbi.nlm.nih.gov
Queue: sup...@open-bio.org
Subject: Changes to NCBI BLAST and E-utilities.
Owner: Nobody
Requestors: mcgi...@ncbi.nlm.nih.gov
Status: new
Ticket <URL: https://helpdesk.open-bio.org/rt/Ticket/Display.html?id=889 >

Dear Colleague:

There are two changes I'd like to make you aware of.

As you may or may not have noticed, we have been working on a new C++
version of the BLAST binaries. In the coming months we will be moving
the C++ binaries into prominence and (slowly) phasing out the C toolkit
binaries.

There are many changes not least of which is a move to individual
binaries for each program (blastn, blastp, etc). We are not sure how
many of your users use BioPerl with the BLAST binaries, my understanding
is that many use BioPerl to to remote BLAST. However, there isa change
to the BLAST results in Text and presumably HTML. This could have an
effect on any parsers which scrape these formats and do not use XML. For
obvious reason, we want to support only the XML format for parsing, but
we thought we should give you heads up on this.
blast 2.2.22

Query: 3307 ------------------------------------------------------------
3307

Sbjct: 390 GSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGP 449

blast 2.2.22+

Query ------------------------------------------------------------

Sbjct 390
GSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGPEAFRGSGP 449

A single line of gaps lacks the Query numbering in the blast+ output.
The C version of blast has numbering in this case. Sample alignment
shown below. According to users the blast+ output without the numbering
breaks bioperl parsers. Wehave heard forma few but I think they may be
older parsers?
The second issue is a policy concerning E-utilities. This was announced
on the utilities...@ncbi.nlm.nih.gov mail-list but you may not
have seen it.

As part of an ongoing effort to ensure efficient access to the Entrez
Utilities (E-utilities) by all users, NCBI has decided to change the
usage policy for the E-utilities effective June 1, 2010. Effective on
June 1, 2010, all E-utility requests, either using standard URLs or
SOAP, must contain non-null values for both the &tool and &email
parameters. Any E-utility request made after June 1, 2010 that does not
contain values for both parameters will return an error explaining that
these parameters must be included in E-utility requests.

The value of the &tool parameter should be a URI-safe string that is the
name of the software package, script or web page producing the E-utility
request.

The value of the &email parameter should be a valid e-mail address for
the appropriate contact person or group responsible for maintaining the
tool producing the E-utility request.

NCBI uses these parameters to contact users whose use of the E-utilities
violates the standard usage policies described
athttp://eutils.ncbi.nlm.nih.gov/entrez/query/static/eutils_help.html#UserSystemRequirements.
These usage policies are designed to prevent excessive requests from a
small group of users from reducing or eliminating the wider community's
access to the E-utilities. NCBI will attempt to contact a user at the
e-mail address provided in the &email parameter prior to blocking access
to the E-utilities.

NCBI realizes that this policy change will require many of our users to
change their code. Based on past experience, we anticipate that most of
our users should be able to make the necessary changes before the June
1, 2010 deadline. If you have any concerns about making these changes by
that date, or if you have any questions about these policies, please
contact eutil...@ncbi.nlm.nih.gov.

Thank you for your understanding and cooperation in helping us continue
to deliver a reliable and efficient web service.

I think you already adhere to this policy but should a user's script not
meet these requirements, than the script will fail and requests will be
turned away with an error message.

Scott D. McGinnis M.A.
NCBI/NLM/NIH
45 Center Drive, MSC 6511
Bldg 45, Room 4AN.44C
Bethesda, MD 20892
mcgi...@ncbi.nlm.nih.gov

_______________________________________________
Bioperl-l mailing list
Biop...@lists.open-bio.org
http://lists.open-bio.org/mailman/listinfo/bioperl-l

[Chris Fields]

Thanks Mauricio. I think the only thing of impact will be the BLAST+ stuff (and Mark has developed a wrapper for it). We'll need to update the SearchIO parser, so any BLAST+ output examples would be tremendously helpful.

chris

[Mark A. Jensen]

all -- note too that blast+ has remote searches built in, and you can
access this functionality in StandAloneBlastPlus by specifying
-remote => 1 to the factory-- cheers MAJ