[sbml-discuss] SBML L2v2 specification vote #4: References to controlled vocabularies

Stefan Hoops shoops at vbi.vt.edu
Thu Dec 22 13:12:07 PST 2005


Hello Howard,

This is not a mood point. The sboTerm DOES NOT replace the MathML. The
problem is that some people including you would like to do that, but we
have been with a fixed set of kinetic laws in level 1 and the people who
had been around that long are all glad that we have left that behind
with all its restrictions. 

The developers do not have the freedom to use either one. Since one is
authoritative (math) and the other optional (sboTerm), i.e., any
software claiming to be SBML conform, compliant, or supporting must
read the math element. However, it might choose to ignore the sboTerm
without negatively affecting its compliance.


Thanks,
Stefan


 On Thu, 22 Dec 2005 11:22:20 -0600
Howard Salis <salis at cems.umn.edu> wrote:

> Stefan,
>     If you only read the sboTerms of the kineticLaw, species, and 
> kinetic parameters you get all the information about the rate law you 
> need. Of course, if you don't parse the MathML expression then you
> can't  compare it to the sboTerms. Whether the sboTerms 'augment' it
> or are a  complementary alternative is a moot point. Both should exist
> to give  developers the opportunity to use _either one_.  And if the
> standard  says the MathML is authoritative...sure, why not. But my
> program won't  read in the SBML model without the sboTerms because
> it's impractical  otherwise. O well.
> 
> -Howard
> 
> >Hello Howard,
> >
> >On Wed, 21 Dec 2005 17:44:08 -0600
> >Howard Salis <salis at cems.umn.edu> wrote:
> >
> >  
> >
> >>This is true. So when an SBML writer changes a reaction in the model
> >>to  another sboTerm, it will also require changing the sboTerms of
> >>each  kinetic parameter and species of that reaction. This is pretty
> >
> >>straightforward from a programming point of view.
> >>
> >>Imho, working with sboTerms vs. MathML expressions is so much easier
> >
> >>from a programming point of view.
> >>    
> >>
> >
> >This is where I think there is a big misunderstanding. You are
> >wrong in assuming that the sboTerm does replace the MathML, it only
> >augments it. If you just have an sboTerm you actually do not have
> >anything. The MathML is always authoritative. If it is missing
> >you have incorrect SBML. Having an sboTerm does not replace reading
> >the MathML. You will always have to do that and check for conformity.
> >If it is conform you may interpret the kinetic law according to your
> >programs needs. If not you must inform the user.
> >
> >Thanks,
> >Stefan
> > 
> >  
> >
> >>-Howard Salis
> >>
> >>
> >>    
> >>
> >
> >  
> >


-- 
Stefan Hoops, Ph.D.
Senior Project Associate
Virginia Bioinformatics Institute - 0477
Virginia Tech
Bioinformatics Facility I
Blacksburg, Va 24061, USA

Phone: (540) 231-1799
Fax:   (540) 231-2606
Email: shoops at vbi.vt.edu


More information about the Sbml-discuss mailing list