Bioperl: XML/BioPerl - design proposal

[Wayne Parrott]

"David J. States" wrote:
> 
> -----Original Message-----
> From:   Wayne Parrott [SMTP:wayne@workingobjects.com]
> ...One of the recent post in this thread caught my attention. It
> was regarding a plan to transform application output into XML, such as
> Blast. I agree this is the way to go. But before "just going and doing
> it" allow me to propose a design model for consideration....
> 
> The discussion of frameworks and design patterns is interesting but under
> emphasizes a critical point, which is the need to develop and maintain
> standards.  
>
> We have a reasonable proposal for sequence XML definition,
> BIOML, as well has a highly developed set of ASN.1 definitions developed
> that NCBI and widely used flatfile definitions for GenBank and
> EMBL/Swissprot.  The NCBI definitions include sequence alignment as well a
> sequence definitions.  Finally, bioperl itself has a well developed set of
> sequence and alignment structures.  To be useful to the community, a new
> set of JAVA tools must support the existing data standards or least provide
> a straightforward way of mapping from existing standards to the standards
> used by the tools.  In all the frameworks discussion, I did not see much
> mention of this.
> 
> David States

David:

Thanks for your comments. As I read your reply the design
principal "separate mechanism from policy" came to mind. The
goal of my framework posting was to focus on "mechanism", a
framework design for translating data/docs from one form to
another, such as XML. Your concern that emerging tools
support existing standards is "policy", i.e., good policy.
The frameworks I've been developing don't ignore this
requirement to support existing standards; I just didn't
mention it in my post. One of the features of the design I
presented is the ability to support mapping between public
and proprietary standards.

In addition to discussing a parsing framework design, I
mentioned BlastXML which proposes an XML form for Blast2
results. This list appears to be a good place to get
constructive criticism on such proposals. In the absence of
an existing schema for Blast I created one. I have some
questions about the DTD design which I'll introduce in a new
thread to this group. I'm assuming that DTD discussions are
in the scope of list. If not, say so.

Wayne Parrott
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc:
http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
====================================================================