Bioperl-guts: seq reimplementation
Sat, 21 Aug 1999 01:01:37 +0100 (BST)
On Fri, 20 Aug 1999, Lovitt, Corey -PMC-CA wrote:
> Sorry to be late in the discussion of the Bio::Seq reworking.
> Has by chance, anyone brought up the notion of making the Bio::Seq class an
> abstract class and implementing concrete subclasses of it such as DNASeq,
> AASeq, RNASeq etc. Common functionality could be stuck into the base class
> and the specialized methods such as 'translate', and 'reverse_translate'
> would have natural locations within the hierarchy. In addition, one could
> associate legal alphabets with each subclass if you chose.
Yup. This has been discussed. I sort of rejected it for two reasons:
a) it changes alot of the semantics from the old seq object
b) we've done fine with a single object for a long time and
I think the main argument to doing it is to "make the object model look
nice", which is a good thing, but not necessary.
I can imagine in the future inhereiting off the Seq object in to Protein
and DNA classes.
> This would throw the onus on the caller to instantiate the correct sequence
> object (potential draw back for integration with the existing code base).
> However, it would likely simplify the Bio::Seq code overall.
more objects... we are only really talking about the complement and
translate functions. Not that much simplification.
I think we are cool with this set up, though I understand your idea and
> -----Original Message-----
> From: Ewan Birney [mailto:email@example.com]
> Sent: Saturday, August 14, 1999 11:16 AM
> To: bioperl guts
> Subject: Bioperl-guts: seq reimplementation
> A final proposal is up at
> I am going to start to get to work on the basis of this proposal,
> assumming that people are ok with it (and only minor changes).
> Notice I need some ideas about
> - case of sequence characters (so petty! and yet so important!)
> - alphabet paranoia
> I have encorporated many ideas given to me about this. Notice
> the backward compatibility functions which throw warnings, and
> the lack of translate but a descriptive error message.
> Ewan Birney
> =========== Bioperl Project Mailing List Message Footer =======
> Project URL: http://bio.perl.org
> For info about how to (un)subscribe, where messages are archived, etc:
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org
For info about how to (un)subscribe, where messages are archived, etc: