[BioPython] Zerg BLAST parser

Hi,

A colleague pointed me to the Zerg BLAST parser 
(http://bioinfo.iq.usp.br/zerg/) which is apparently pretty quick. It only 
had a wrapper for Perl, so I knocked one up for Python, partly as a 
learning exercise in Pyrex. If you think you'll find it useful, or would 
like to improve on (and/or replace) it, you can find it at 
http://bioinf.scri.sari.ac.uk/lp/pyzerg.shtml

Have fun,

Dr Leighton Pritchard AMRSC
PPI, Scottish Crop Research Institute
Invergowrie, Dundee, DD2 5DA, Scotland, UK
L.Pritchard <at> scri.sari.ac.uk
PGP key 47B4A485: http://www.keyserver.net http://pgp.mit.edu

_______________________________________________
BioPython mailing list  -  BioPython <at> biopython.org
http://biopython.org/mailman/listinfo/biopython

Re: [BioPython] Zerg BLAST parser

Hi Leighton;

> A colleague pointed me to the Zerg BLAST parser 
> (http://bioinfo.iq.usp.br/zerg/) which is apparently pretty quick. It only 
> had a wrapper for Perl, so I knocked one up for Python, partly as a 
> learning exercise in Pyrex. If you think you'll find it useful, or would 
> like to improve on (and/or replace) it, you can find it at 
> http://bioinf.scri.sari.ac.uk/lp/pyzerg.shtml

Cool! Thanks for doing this and posting it. I added the link to the
ScriptCentral page (http://www.biopython.org/scriptcentral/). Thanks
again!

Brad
_______________________________________________
BioPython mailing list  -  BioPython <at> biopython.org
http://biopython.org/mailman/listinfo/biopython

[BioPython] Zerg BLAST parser - timings

Hi,

While I don't imagine Zerg will necessarily be the best option for all 
circumstances, running time trials with the script below (extracts all 
query sequence names from NCBI standalone BLAST output) gives on our Linux 
server:

361kb output file, 3 runs:
-------------------------------------------
Module                  Mean            StDev

NCBIStandalone.Parser   0.986s          0.003s

Zerg                            0.032s          0.005s

367Mb output file, 3 runs:
-------------------------------------------
Module                  Mean            StDev

NCBIStandalone.Parser   1461.08s        20.95s

Zerg                            58.29s          4.38s

There are almost certainly faster ways to use the Biopython parser code 
(specialist consumer, perhaps?) than in the test script given below, but I 
couldn't think of them nearly as quickly as I could implement the quick 
run-though with the Zerg parser.

In this simple case, the Zerg code turned out to be between 25 and 30 times 
faster than the Biopython code in the script below.  The Zerg wrapper and 

Re: [BioPython] Zerg BLAST parser

At 11:32 01/09/2003 -0400, you wrote:
> > [...] the Zerg BLAST parser (http://bioinfo.iq.usp.br/zerg/) [...] only
> > had a wrapper for Perl, so I knocked one up for Python [...] at
> > http://bioinf.scri.sari.ac.uk/lp/pyzerg.shtml
>
>Cool!

Wow, I got a 'Cool' from Brad! - that's the second really good thing to 
happen to me today ;)

>Thanks for doing this and posting it. I added the link to the
>ScriptCentral page (http://www.biopython.org/scriptcentral/).

And that's the third...

>Thanks again!

No problem - so long as people let me know under what circumstances it 
stops working, and where my docs are *really* unhelpful so I can fix them.

On a tangent, would anyone be interested in a publication-quality genome 
schematic (linear and circular genome diagram, with labels, scales and 
plots) drawing module?  I have one that would probably benefit from public 
release and comment.

Dr Leighton Pritchard AMRSC
PPI, Scottish Crop Research Institute
Invergowrie, Dundee, DD2 5DA, Scotland, UK
L.Pritchard <at> scri.sari.ac.uk
PGP key 47B4A485: http://www.keyserver.net http://pgp.mit.edu

Re: [BioPython] Zerg BLAST parser

Leighton Pritchard wrote:
>> > [...] the Zerg BLAST parser (http://bioinfo.iq.usp.br/zerg/) [...] only
>> > had a wrapper for Perl, so I knocked one up for Python [...] at
>> > http://bioinf.scri.sari.ac.uk/lp/pyzerg.shtml

Thanks!

A minor note (I don't know Pyrex): can Zerg constants, like QUERY_NAME, 
be retained in pyzerg?

[...]

> 
> On a tangent, would anyone be interested in a publication-quality genome 
> schematic (linear and circular genome diagram, with labels, scales and 
> plots) drawing module?  I have one that would probably benefit from 
> public release and comment.

You bet!

> 
> 
> Dr Leighton Pritchard AMRSC
> PPI, Scottish Crop Research Institute
> Invergowrie, Dundee, DD2 5DA, Scotland, UK
> L.Pritchard <at> scri.sari.ac.uk
> PGP key 47B4A485: http://www.keyserver.net http://pgp.mit.edu
> 

yves

Re: [BioPython] Zerg BLAST parser

At 09:42 02/09/2003 +0200, Yves Bastide wrote:
>Leighton Pritchard wrote:
>>> > [...] the Zerg BLAST parser (http://bioinfo.iq.usp.br/zerg/) [...] only
>>> > had a wrapper for Perl, so I knocked one up for Python [...] at
>>> > http://bioinf.scri.sari.ac.uk/lp/pyzerg.shtml
>
>Thanks!
>
>A minor note (I don't know Pyrex): can Zerg constants, like QUERY_NAME, be 
>retained in pyzerg?

Yes, and I've added this, now. You have to address them as zerg.QUERY_NAME 
and so on, but it works.

Dr Leighton Pritchard AMRSC
PPI, Scottish Crop Research Institute
Invergowrie, Dundee, DD2 5DA, Scotland, UK
L.Pritchard <at> scri.sari.ac.uk
PGP key 47B4A485: http://www.keyserver.net http://pgp.mit.edu

_______________________________________________
BioPython mailing list  -  BioPython <at> biopython.org
http://biopython.org/mailman/listinfo/biopython

Re: [BioPython] blast parser slows down under python2.3

I take it that you are applying these patches in CVS?
I have only downloaded the tarball for BioPython, would you suggest i 
check it out from CVS and what tag should I use?

Andy

On Sunday, August 31, 2003, at 05:44 PM, Jeffrey Chang wrote:

> I have applied the patch.  Thanks very much!
>
> The regression tests now work again.  For the tests that print out 
> booleans, I am now explicitly printing out 0 or 1, for backwards 
> compatibility.
>
> I have also gone through and changed some more instances of apply to 
> the new call syntax.  Please let me know if there appears to be any 
> problems.
>
> Jeff
>
>
>
> On Friday, August 29, 2003, at 12:07  PM, Jeffrey Chang wrote:
>
>> Hey, thanks very much for the note, and the patch (mailed separately).
>>
>> Python 2.3 also seems to have broken some of the regression tests.  
>> The boolean type gets printed out as "True" and "False" rather than 1 
>> or 0 as before.
>>

[BioPython] care for an rpm spec file?

attached is an rpm spec file to build biopython 1.21,
it has builddependencies setup for mandrake 9.1. The
chatter about "Martel" and "Reportlab" is disabled
with patching setup.py - sadly, that seems necessary.

btw, please fix that annoying martel install bug.
While running setup, a missing "Martel" dependency
is claimed - and the default is "NO", do not continue
with installation, i.e. a missing required dependency.

As a search through the mailinglist shows, this is
simply wrong - however an innocent user will not know
this since the "installation help" does not have a
hint about a Martel installation dependency.

Please add a paragraph on
http://biopython.org/docs/install/Installation.html
in section 3 - tell about what "Martel" is and that
it is okay to continue. Otherwise, we should follow
the setup.py recommendation, and do _not_ continue.

-- have fun, guido    http://google.de/search?q=guidod

%define name biopython
%define version 1.21
%define release py2.2_1mdk

Summary: BioPython
Name: %{name}

Re: [BioPython] blast parser slows down under python2.3

I was wondering if you could patch the stable branch?  Reason is that 
some of what I do may turn into production scripts and I would rather 
not have it tied to bleeding edge code..
Thanks

On Tuesday, September 2, 2003, at 12:45 PM, Jeffrey Chang wrote:

> Yep, it's patched against the CVS.  There is no tag -- just the most 
> recent version.  We've only been tagging the releases.
>
> Jeff
>
>
> On Tuesday, September 2, 2003, at 07:30  AM, Andrew Nunberg wrote:
>
>> I take it that you are applying these patches in CVS?
>> I have only downloaded the tarball for BioPython, would you suggest i 
>> check it out from CVS and what tag should I use?
>>
>> Andy
>>
>> On Sunday, August 31, 2003, at 05:44 PM, Jeffrey Chang wrote:
>>
>>> I have applied the patch.  Thanks very much!
>>>
>>> The regression tests now work again.  For the tests that print out 
>>> booleans, I am now explicitly printing out 0 or 1, for backwards 
>>> compatibility.
>>>
>>> I have also gone through and changed some more instances of apply to 

[BioPython] Re: streamerror_in_local_blast.py

Could you please provide the entire traceback? It's pretty hard to find
an error without it ;)

Thanks,

./I

kim750 <at> ggbio.com wrote:
> Hi!
> My name is Yeon-Ki.  I am a novice with biopython.
> Now I am tring to use local_blast.py.  But the moment i typed local_blast.py I have Running blastall...
> Traceback (most recent call last):  and 
> whole bunch of "Bio/Blast/NCBIStandalone.py" errors and SyntaxError: Unexpected end of stream at the
end of error message.
> 
> I down-loaded  a current version of  NCBIStandalone.py --> Revision 1.49 / (as text) / (view) - annotate -
[select for diffs] , Tue Sep 2 17:47:29 2003 UTC (16 hours, 11 minutes ago) by jchang.  I replaced the old one
in the Bio/Blast/NCBIStandalone.py with new one , but still I could not use the local_blast.py.
> 
> Could you help me out?
> 
> Thank you.
> Yeon-Ki
> 
> 
>  
> 
> 
> Yeon-Ki Kim, Ph.D.
> Research Scientist