[Bioperl-l] urgent help required - syntax for using paramatersdifferent from default in standalone blast

[Chris Fields]

On Mar 13, 2006, at 1:39 AM, neeti somaiya wrote:

> Hi,
>
> I am using blast 2.2.13 on a linux machine.
>
> I tried this:
>
> @params = ('program'  => 'blastn','database' => 'human.rna.fna','e' =>
> '0.0001', 'Matrix' => 'BLOSUM80', '_READMETHOD' => 'SearchIO' );
>
> But, this din't help.
>
> I checked the data folder, it has BLOSUM80.
>
> Even a simple blastall command like
> blastall -i tryn.fasta -p blastn -d human.rna.fna -o blastoutput -M  
> BLOSUM80
> doesnt show BLOSUM80 in the results.

This should tell you something here and is likely the source of the  
problem.  Do you have the .ncbirc file set properly?

 From BLAST install:

2) Create a .ncbirc file. In order for Standalone BLAST to operate, you
have will need to have a .ncbirc file that contains the following lines:

[NCBI]
Data="path/data/"

Where "path/data/" is the path to the location of the Standalone BLAST
"data" subdirectory. For Example:

Data=/root/blast/data

The data subdirectory should automatically appear in the directory where
the downloaded file was extracted. Please note that in many cases it may
be necessary to delimit the entire path including the machine name and
or the net work you are located on. Your systems administrator can help
you if you do not know the entire path to the data subdirectory.

Make sure that your .ncbirc file is either in the directory that you
call the Standalone BLAST program from or in your root directory.

I believe this can be placed in your home directory though I can't  
test that out.  Are your databases, matrices, etc all in the data  
directory?


> What should I do?
>

Don't panic!


> Any other approach to change the parameters?
> Why is it that e value is taken, but not any other parameter?

These are the ones I found in StandAloneBlast, just by looking (hint  
hint):

	@BLASTALL_PARAMS = qw( p d i e m o F G E X I q r v b f g Q
			    D a O J M W z K L Y S T l U y Z);

Chris


>
>
>
> On 3/12/06, Chris Fields <cjfields at uiuc.edu> wrote:
>>
>> Ah.  I thought that RemoteBlast and StandAloneBlast had similar
>> settings.  My bad.  Would be nice to maybe move some of the common
>> features to another module that both can inherit from (like parsing/
>> saving output, map parameters for each like SearchIO, etc).
>>
>> Chris
>>
>> On Mar 11, 2006, at 4:17 PM, Torsten Seemann wrote:
>>
>>>> Although I believe there is a major reworking
>>>> of the Bio::Tools::Run BLAST modules planned
>>>
>>> Yes there is, but Roger Hall fell ill during our initial
>>> discussions so we are
>>> a little behind on that one.
>>>
>>> I'm pretty sure that
>>>> everything still works, at least in the latest developer version.
>>>> Lots of people are running it so I'm guessing something is wrong  
>>>> with
>>>> the logic here otherwise we would have heard about this a while  
>>>> ago.
>>>
>>> Agreed - at least in the bioperl-live version.
>>>
>>>> I'm not an expert about this module either but I believe Brian's
>>>> right about _READMETHOD.   When using SearchIO directly this is set
>>>> using "-readmethod => 'SearchIO'" for text, 'blastxml' for xml, and
>>>> 'blasttable' for tabular.
>>>
>>> Actually, 'blastxml' is NOT supported in bioperl-live version, and
>>> it doesn't
>>> appear to have ever been supported! It won't be hard to patch in.
>>>
>>> Also, the current StandAloneBlast does not handle these newer
>>> "blastall"
>>> parameters (need to add to @BLASTALL_PARAMS)
>>>
>>>    -R  PSI-TBLASTN checkpoint file [File In]  Optional
>>>    -n  MegaBlast search [T/F]
>>>    -L  Location on query sequence [String]  Optional
>>>    -A  Multiple Hits window size, default if zero (blastn/megablast
>>> 0, all
>>> others 40 [Integer]
>>>    -w  Frame shift penalty (OOF algorithm for blastx) [Integer]
>>>    -t  Length of the largest intron allowed in a translated
>>> nucleotide sequence
>>> when linking multiple distinct alignments. (0 invokes default
>>> behavior; a
>>> negative value disables linking.) [Integer]
>>>    -B  Number of concatenated queries, for blastn and tblastn
>>> [Integer]  Optional
>>>    -V  Force use of old engine [T/F]  Optional
>>>    -C  Use composition-based statistics for tblastn:
>>>    -s  Compute locally optimal Smith-Waterman alignments (This
>>> option is only
>>>
>>> According to the POD for StandAloneBlast
>>>> SearchIO::blast is default;  have you tried not using that flag
>>>> (removing it)?
>>>
>>> Yes, _READMETHOD='BLAST' (which uses Bio::SearchIO for parsing) is
>>> the default,
>>> and does not need to be set in Neeti's examples.
>>>
>>> --
>>> Torsten Seemann
>>> Victorian Bioinformatics Consortium, Monash University, Australia
>>> http://www.vicbioinformatics.com/
>>> Phone: +61 3 9905 9010
>>> _______________________________________________
>>> Bioperl-l mailing list
>>> Bioperl-l at lists.open-bio.org
>>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>
>> Christopher Fields
>> Postdoctoral Researcher
>> Lab of Dr. Robert Switzer
>> Dept of Biochemistry
>> University of Illinois Urbana-Champaign
>>
>>
>>
>>
>
>
> --
> -Neeti
> Even my blood says, B positive
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l

Christopher Fields
Postdoctoral Researcher
Lab of Dr. Robert Switzer
Dept of Biochemistry
University of Illinois Urbana-Champaign