[Bioperl-l] new Bio::Restriction classes

[Heikki Lehvaslaiho]

Rob Edwards and I have been writing new restriction analysis
classes. These will eventually replace the long serving
Bio::Tools::RestrictionEnzyme by Steve Chervitz. The first working
versions are in CVS.

Please send your comments and suggestions to the list.

There is an UML graph at http://bio.perl.org/images/bio_restriction.png.
A more verbose overview is below:

- The name space is Bio::Restriction

  Restriction analysis is such a basic and important part of  
  molecular biology and bioinformatics that giving it its own top 
  level name space is reasonable.

- Bio::Restriction::Enzyme class knows (almost) everything there to
  know about restriction enzymes. 

  There are two subclasses of Enzyme:
  -- Bio::Restriction::Enzyme::MultiSite
  -- Bio::Restriction::Enzyme::MultiCut

  that handle relatively rare cases when recognition sites of the
  enzyme is more complex than can be expressed using IUPAC code or 
  there are four cut sites per site.


- Bio::Restriction::EnzymeCollection object is a set of Enzymes. It is
  created by Bio::Restriction::IO. This class works like Bio::SeqIO
  class, but if you call it without parameters, you get a default
  selection of 530 bare-boned Enzymes.

  If you want select your own set, you can read in data from REBASE
  database flat files. Two formats, itype2 and withrefm, are now
  supported.

  Any attribute of an Enzyme can be used to filter them out of one
  collection into an other.

- Bio::Restriction::Analysis uses an EnzymeCollection to cut any
  Bio::PrimarySeqI implementing sequence into fragments. There are
  methods to find enzymes that cut the sequence to a given number of
  time, or do not cut at all, and retrieve the fragments.


Enjoy,

	-Heikki & Rob