[BioPython] Blast 2.2 or better

[Osborne, John]

Hi Jeff,

Thanks for your quick response, I took a look at the biopython blast
code and it appears to me that the problem is that the blast result
description headers are slightly different in the version of blast I am
using.  As it turns out I now recognize the results I gave it to parse
as being from mpiblast 1.1 wherein lies the rub... mpiblast gives the
first sequence description of a hit prefixed with a '>' and if there are
identical hits it just lists them out after the initial one with no
leading '>'.  I think this is causing the problem, scan_alignments does
look for a '>' and start to process an alignment which is not there.

I've included the blast output and the error file, I have tried
debugging this myself but I have run into some odd behaviour.  I tried
adding a print statement (print "boo") to NCBIStandalone.py and as
output I got 'print "boo"' ie) it included the word print and quotes?!
I just started with python last week and found this odd to say the
least.  I also am also confused by read_and_call functions?  Is this
biopython or python?  I tried grepping for "def read_and_call" in the
biopython directory but can't find it, how would I go about finding this
(and similar) function?

Any help appreciated for a potential python convert,

 -John


I


-----Original Message-----
From: Jeffrey Chang [mailto:jchang at jeffchang.com] 
Sent: Sunday, November 16, 2003 9:18 PM
To: Osborne, John
Subject: Re: [BioPython] Blast 2.2 or better

Hi John,

The parser was started on Blast 2.0, but we have tried to keep it up to 
date with Blast.  It does work with some flavors of Blast 2.2, but 
unfortunately, the format changes very fast.  Sometimes it changes 
faster than the parser.  So, yes, a patch that updates it with your 
current version is best.  Next best would be to send email to the list, 
with a description of the problem and a sample of the output that 
breaks the parser.  Good luck!

Jeff


On Nov 16, 2003, at 8:31 PM, Osborne, John wrote:

> Hi,
>
> I just started learning python last week, hoping to use some general
> purpose blast parsing code with little effort on my part and I ran
into
> trouble using the standalone blast parser.  It looks like it was 
> written
> for only blast 2.0 about 3 or 4 years ago.  Is this an area that needs
> contribution or is there something else I should be using other than
> this for fast blast parsing?  I'm using NCBI's 2.2 blast locally.
>
>
>
>  -John
>
>
>
> _______________________________________________
> BioPython mailing list  -  BioPython at biopython.org
> http://biopython.org/mailman/listinfo/biopython

-------------- next part --------------
A non-text attachment was scrubbed...
Name: error
Type: application/octet-stream
Size: 1943 bytes
Desc: error
Url : http://portal.open-bio.org/pipermail/biopython/attachments/20031117/aa86cef8/error-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: oneresult
Type: application/octet-stream
Size: 143825 bytes
Desc: oneresult
Url : http://portal.open-bio.org/pipermail/biopython/attachments/20031117/aa86cef8/oneresult-0001.obj