[Bioperl-l] Bio::Tools::BPlite

Ewan Birney birney@ebi.ac.uk
Sun, 16 Jul 2000 11:32:45 +0000 (GMT)


On Fri, 14 Jul 2000, Lorenz Pollak wrote:

> Hi to everyone!
> 
> I have just commited the first running version
> of the ported BPlite module...
> 

Great!

> It is lacking of documentation and testing scripts
> but I will add this stuff in the next days.
> 
> It consists of three files:
> Bio/Tools/BPlite.pm
> Bio/Tools/BPlite/Sbjct.pm
> Bio/Tools/BPlite/HSP.pm
> 
> The main thing about it is that the HSP class is 
> inherited from Bio::SeqFeature::FeaturePair.
> But I am not really sure, if this is of use
> for anyone, because the only advantage would be,
> that you could use SeqFeatureI and RangeI methods
> on "start" and "end" of the two aligned sequences.
> 

There is a massive advantage here, as alot of the systems in
Bio::Seq/Bio::SeqIO rely on things inherieting from Bio::SeqFeatureI - you
should also set primary_tag and source_tag to sensible values.

The RangeI inheritence gives us the ability to say

   if( $hsp->overlaps($exon) ) {
      # then do something
   }

Finally it is likely that we need to add a to_FTHelper function to the HSP
module to make it play nicely with the EMBL/GenBank IO streams. This might
be one for me.


Many, many thanks lorenz. 


> The rest of information (sequences,scores) is
> only available via the HSP object...
> 
> Any other suggestions for changes or additions 
> are always welcome!
> 



> kind regards,
> Lorenz
> 
> PS: I tested it against t/blast.report and it
> first failed, because it didn't expect a line
> starting with "CPU time: ..." at the end of
> the alignments. So I added this and then it
> worked.

Have you added a BPLite specific test?


> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
>