[Pipet Devel] BioML vs BSML

[Harry Mangalam]

Hi Jeff et al,

As you probably have found out now, BioML is being used by the bio.perl
group and the perl masters at perl.org already have a pretty large archive of
useful scripts for manipulating XML.

And it looks like, in a VERY fast spin thru the docs, that using their XML
parser tools, it may be possible to use both of these XMLs, using these perl
modules to handle parsing and interconversion.

ie:
http://www.perl.com/pace/pub/perldocs/1998/11/xml.html
http://www.perl.com/pace/pub/perldocs/1998/12/cooper-01.html

perl also has gtk bindings and in general, I've heard about and done more
things using perl in the bio world than with python.  Not to say python
doesn't make some or even most things easier - just that perl has a proven
track record in the bio area.

It seems not to be at cross purposes to the objectives of LOCI to implement
chunks of it in perl, no?  As long as it remains easliy implementable, and
usable, and freely re-distributable, perl is as much of an option as
python, isn't it?

Also, how is the LOCI project planning on handling the display of the
results of this effort?  Both the BioML and the BSML browsers that are
available are MS-centric and certainly do not use gtk.  The only browser
that I'm aware of that does is gzilla (www.gzilla.com) and mnemonic (
http://www.mnemonic.org/).  Are you planning on using either of these for
your display dev platform?  Or are you not implementing any display
technology?

VisualGenomics is planning a Java-based BSML browser, but I'm sure that it
won't use gtk, unless heavily funded.  It'll probably be written with the
swing classes - what's the redist controls on swing - I'm not a Java
follower - sorry.

OK - enough mind rot from me for the present...

Cheers
harry

On Sat, 23 Jan 1999, J.W. Bizzaro wrote:

> Harry,
> 
> We had some rather lengthy exchanges about the use of BSML and CML in Loci.  We
> were looking for an XML that would work well for both sequence and structure
> information.  But it seemed that neither was good at both.
> 
> In sort, we stepped back from the issue, and decided we would try to support
> BOTH, until something better came along.  We even considered making our own XML
> (we would call it BMML, "BioMolecule Markup Language").  What we really wanted
> was just one language that would give us an excellent description of large bio
> macromolecules.  Well, of course that would have been too much to take on right
> now.
> 
> Konrad actually has a lot to say about descriptive languages for macromolecule
> structure.  He had corresponded with Peter Murray Rust, the author of CML and
> someone rather influential in the development of XML.  Konrad, as he can tell
> you, is not satisfied with any current format.
> 
> BioML comes as a bit of a surprise to me.  It seems to be brand new.  Looking it
> over a bit, it does seem to do a better job at handling structure than BSML, and
> I like the inclusion of all sorts of biological information (it can describe
> organisms as well, it seems)...although some may argue this is "bloat".
> 
> I would like Konrad to give us his impression of BioML.  It would be nice to use
> one XML rather than two.  My big question, however, is the licensing.  The Web
> page says to contact David Fenyo about the "commercial" use of BioML.  I wonder
> if this is one of those "well, as long as you don't make money on it" licenses. 
> If so, we have problems: It won't fit with GNU GPL.
> 
> I wrote a message to David Fenyo about this, and to see if he can give us a
> contrast between BioML and BSML.  A cc will be sent to the Tulip list.
> 
> 
> Cheers!
> Jeff
> bizzaro at bc.edu
> 
> 
> Harry Mangalam wrote:
> > 
> > Hi All,
> > 
> >    Having only recently come to this arena, what's the group's evaluation of
> > the relative merits of BSML:
> > http://www.visualgenomics.com/sbir/rfc.htm
> > 
> > vs BioML:
> > http://www.proteometrics.com/BIOML/
> > 
> > I'm still going thru the text of the specs but if any of you have strong
> > arguments regarding either approaches, I'd very much appreciate it.  The
> > bioperl people seem to like the BioML approach.
> > 
> > Also, if you have come to an approach that can rationalize the competing
> > stds, please let me know.
> > 
> > Cheers
> > Harry
> 
> -- 
> J.W. Bizzaro                  Phone: 617-552-3905
> Boston College                mailto:bizzaro at bc.edu
> Department of Chemistry       http://www.uml.edu/Dept/Chem/Bizzaro/
> --
> 

Cheers,
Harry

Harry J Mangalam, Developmental + Cell Biology
Rm 4201, Biological Sciences II, UC Irvine, Irvine, CA, 92697
(949) 824 4824[vox], (949) 824 8551[fax], mangalam at uci.edu
http://hornet.bio.uci.edu/~hjm/