[Bioperl-l] allowgaps parameter in Bio::SearchIO

[Jason Stajich]

It's not parsed separately probably.  What you're seeing is a default
parameter which is reset to no when parsing WU-BLAST output (which
conviently has a line which says 'nogaps' to make this determination
easily).

The logic of the parsing will have to be reconsidered to try and add this
sort of information in I guess.  Unfortunately NCBI-BLAST doesn't do a
particularly nice job of reporting what were all the input parameters to
the BLAST run so it a bit of work to try and infer all of them from the
output based on what is present/absent.

A bit more work and attention needs to be paid to the stats and param
parsing, as we are not really capturing all of them.  Something I'd love
for someone to try and help with, or I will try and get to it towards the
end of November when my schedule frees up a little bit.

-jason

On Fri, 31 Oct 2003, Daniel Lang wrote:

> Hi,
> I´m getting wrong values for the "allowgaps" parameter while parsing
> NCBI-like BLAST results:
> The parser does not detect if a the report was generated with -g F
> switch(ungapped) and the allowgaps values is "yes", when it should be
> "no"...
> I guess the parameter is based on the occurence of "Gapped" after the
> "Lambda"-part, right?
>
> Here is an example BLAST output snipplet(-g F):
>
> Searching..................................................done
>
>   ***** No hits found ******
>
>    Database: ppp_orfpep
>      Posted date:  Sep 18, 2003  6:31 AM
>    Number of letters in database: 2,367,445
>    Number of sequences in database:  14,583
>
> Lambda     K      H
>     0.318    0.135    0.401
>
>
> Matrix: BLOSUM62
> Number of Hits to DB: 378,425
>
> Or did I missinterpret the parameter?
> Thanks ind advance,
> Daniel:)
>
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu