[Bioperl-l] experimental Bio::Search::Tiling implementation

[Mark A. Jensen]

Great, guys -- thanks--
Yesterday I committed a fix (r15689) that took care of problems
with tile_hsps()'s handling of gapped alignments. This may have
solved some of the old problems. I will have a go at the data
Sendu mentions. (I didn't go through all 1093 tests in blast.t;
will look more closely...) Will also try to make representations
of the HSPI methods that are generalizable, and use these to
flesh out the TilingI interface (create aliases back to HSPI, e.g.)
cheers- Mark
----- Original Message ----- 
From: "Chris Fields" <cjfields at illinois.edu>
To: "Sendu Bala" <bix at sendu.me.uk>
Cc: "Steve Chervitz" <sac at bioperl.org>; "BioPerl List" 
<bioperl-l at lists.open-bio.org>; "Mark A. Jensen" <maj at fortinbras.us>
Sent: Tuesday, May 19, 2009 9:14 PM
Subject: Re: [Bioperl-l] experimental Bio::Search::Tiling implementation


> On May 19, 2009, at 7:07 PM, Sendu Bala wrote:
>
>> Steve Chervitz wrote:
>>> Great work. My SearchUtils tiling function has been lingering for far
>>> too long (at least a decade).
>>
>> I actually reworked it a little from revisions r10199 to r11247.
>>
>> Mark: tricky.wublast and frac_problems.blast were the test files  that had 
>> problems that needed solving.
>>
>> The corresponding tests were in SearchIO.t but have since  (hopefully!) been 
>> moved to another test file somewhere.
>
> They should be in t/SearchIO/blast.t; the SearchIO tests were split up  based 
> on format.  Was there a bug linked to these that need to be  merged with 
> Mark's?
>
> Mark: great work!  I suppose the next step is to try refactoring the  relevant 
> methods in HSPI's to see what happens.
>
> chris
>
>
>