[Bioperl-guts-l] [Bug 2637] new modules for HIV sequence queries

[bugzilla-daemon at portal.open-bio.org]

http://bugzilla.open-bio.org/show_bug.cgi?id=2637





------- Comment #10 from maj at fortinbras.us  2008-11-26 10:07 EST -------
Thanks, Heikki-
(In reply to comment #9)
> Mark,
> 
> I had a quick look at your classes (the original tarball).  A few things came
> to mind:
> 
> 1. You know that we have plans of splitting bioperl into smalled chunks.
> Depending how far you'll want to take this, HIV specific stuff might be one of
> them. As they are now, I happy to have have it as part of bioperl-live.

Right-- this is pretty specialized (and has the chance of being torpedoed by
the LANL developers, if they ever overhaul their site, since this is really a
hack in disguise). To have it in the main trunk, however, may increase its use,
and ultimately might prod LANL to provide a real code-accessible backend!

I do think there are virus/infectious disease oriented concepts and
corresponding modules that could be profitably lumped together and packaged,
once a critical mass of code is reached. Could be fun.

> 
> 2. You use LWP::UserAgent directly. You might want to use Bio::WebAgent. You
> get more bioperl-like attributes and fetch delay abilities for free.
> 

Sounds good.

> 
> 3. Using annotations to store the additional information is fine, cause that
> allows you to use normal rich sequence formats, but accessing is not that
> straightforward. Why do not you create a new sequence object that inherits from
> Bio::Seq::RichSeqI (or Bio::Seq::RichSeq) and adds convenience methods for
> accessing the additional information.
> 

I will look at this carefully. [The patches above remove the special
annotations all together in favor of a more fluid and easier to use 'internal'
solution.]


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.