Clarrify the use of notes

Herbert Sauro

unread,

Jan 27, 2006, 2:11:23 PM1/27/06

to sbml-d...@caltech.edu

Can anyone help me clarrify the use of notes in sbml. The spec gives at
least two variants on using notes:

One example has the following form:

<notes>
<body xmlns="http://www.w3.org/1999/xhtml">
Simple branch system.
The reaction looks like this:
reaction-1: X0 -> S1; k1*X0;
reaction-2: S1 -> X1; k2*S1;
reaction-3: S1 -> X2; k3*S1;
</body>
</notes>

While another has the simpler form:

<notes>
<html:p>((vm * s1)/(km + s1))*cell</html:p>
</notes>

Should a program be ready to parse both forms and are there any other
variants?

Herbert SAuro
------------------------------------------------------------------------
---------
Herbert M Sauro
Assistant Professor
Keck Graduate Institute
535 Watson Drive
Claremont, CA, 91711
Tel: 909 607 0377
web site: www.sys-bio.org

Trying to convice a believer of ID that ID isn't science is like trying
to teach a pig to sing...it wastes your time and annoys the pig.

Pedro Mendes

unread,

Jan 27, 2006, 4:15:16 PM1/27/06

to sbml-d...@caltech.edu

This is a good question, Herbert. I don't know the answers either but would
like to know them too. In addition I would ask if a program is required to
be able to process notes other than just read them. For example, if a
program does not display notes will it then be considered non-compliant? I
think that this should not be a problem for compliance, but would like to
know the opinion of others.

Pedro

--
Pedro Mendes
Research Associate Professor
Virginia Bioinformatics Institute,
Virginia Tech, Washington St.,
Blacksburg, VA 24061-0477, USA
http://mendes.vbi.vt.edu fax:+1-540-231-2606

Michael Hucka

unread,

Jan 27, 2006, 7:24:37 PM1/27/06

to SBML Discussion List

mendes> This is a good question, Herbert. I don't know the
mendes> answers either but would like to know them too. In
mendes> addition I would ask if a program is required to
mendes> be able to process notes other than just read
mendes> them. For example, if a program does not display
mendes> notes will it then be considered non-compliant? I
mendes> think that this should not be a problem for
mendes> compliance, but would like to know the opinion of
mendes> others.

It was the intention of the specification as written that
compliance does not require being able to process notes.

Notes are not only optional; even when they're present, they
can have no effect on the substance of the model. They're
purely for human consumption.

I'm sorry this is not more clear in the documentation.

MH

Ben Bornstein

unread,

Jan 27, 2006, 5:13:55 PM1/27/06

to SBML Discussion List

Hi Herbert,

The only difference between the two forms is one contains an XHTML
<body> element and the other does not. The "html" in <html:p> is an
XML namespace prefix. It is *not* an <html> element.

I'm assuming you pulled the second "form" from Example 5.2 (page 34) in
the SBML L2v1 specification. If you look at the start of the SBML,
you'll see the XML namespace declaration that associates the prefix
"html" with the namespace URI "http://www.w3.org/1999/xhtml":

<sbml xmlns="http://www.sbml.org/sbml/level2" level="2"
version="1" xmlns:html="http://www.w3.org/1999/xhtml">

You could rewrite:

as:

<notes>
<html:p xmlns:html="http://www.w3.org/1999/xhtml">

((vm * s1)/(km + s1))*cell</html:p>

</html:p>
</notes>

or even:

<notes>

((vm * s1)/(km + s1))*cell</html:p>

</notes>

All three forms are equivalent. Also, the "html" doesn't matter, it
could just as easily be "foo". What matters is the namespace URI.
Coping with such syntactic variations can be frustrating. That's why
it's best to leave the details to an XML parser like Xerces, Expat, or
libxml (and that's why libSBML relies on an XML parser, instead of
implementing its own).

Ben

Frank Bergmann

unread,

Jan 27, 2006, 5:20:38 PM1/27/06

to SBML Discussion List

I just had a word with Frank Bergmann and he pointed out to me that both
variants are of course html, so there is really no issue, I'm not sure
how I missed that. One has to be aware however that upon encountering a
notes tag, your parser should read the entire contents as a single
block. I should also note that I don't think biomoidels.net uses notes
but uses rdf inside an annotation instead. Does libsbml yet read the
biomodels annotation, I think this would be useful to have. However the
point made by Pedro was interesting, are notes necessary for compliance,
probably not since I think the sbml docs says notes is optional?

Herbert Sauro

-----Original Message-----
From: Pedro Mendes [mailto:men...@vbi.vt.edu]
Sent: Friday, January 27, 2006 1:15 PM
To: sbml-d...@caltech.edu
Subject: Re: Clarrify the use of notes

This is a good question, Herbert. I don't know the answers either but
would like to know them too. In addition I would ask if a program is
required to be able to process notes other than just read them. For
example, if a program does not display notes will it then be considered
non-compliant? I think that this should not be a problem for compliance,
but would like to know the opinion of others.

Pedro

On Friday 27 January 2006 14:11, Herbert Sauro wrote:
> Can anyone help me clarrify the use of notes in sbml. The spec gives
> at least two variants on using notes:
>
> One example has the following form:
>
> <notes>
> <body xmlns="http://www.w3.org/1999/xhtml">
> Simple branch system.
> The reaction looks like this:
> reaction-1: X0 -> S1; k1*X0;
> reaction-2: S1 -> X1; k2*S1;
> reaction-3: S1 -> X2; k3*S1;
> </body>
> </notes>
>
> While another has the simpler form:
>
> <notes>
> <html:p>((vm * s1)/(km + s1))*cell</html:p>
> </notes>
>
> Should a program be ready to parse both forms and are there any other
> variants?
>

> Herbert SAuro
> ----------------------------------------------------------------------

> --
> ---------
> Herbert M Sauro
> Assistant Professor
> Keck Graduate Institute
> 535 Watson Drive
> Claremont, CA, 91711
> Tel: 909 607 0377
> web site: www.sys-bio.org
>
> Trying to convice a believer of ID that ID isn't science is like
> trying to teach a pig to sing...it wastes your time and annoys the
pig.

--
Pedro Mendes
Research Associate Professor
Virginia Bioinformatics Institute,
Virginia Tech, Washington St.,
Blacksburg, VA 24061-0477, USA
http://mendes.vbi.vt.edu fax:+1-540-231-2606

606

Michael Hucka

unread,

Jan 27, 2006, 7:28:24 PM1/27/06

to SBML Discussion List

frank_bergmann> Does libsbml yet read the biomodels
frank_bergmann> annotation, I think this would be useful
frank_bergmann> to have.

LibSBML does not currently provide special support for
BioModels DB (or any other) annotations. The general goal
in that respect is to provide an API to work with RDF inside
annotations; this would allow other annotations to be read
too. It's definitely on the short-list of features to
implement.

MH

Michael Hucka

unread,

Jan 27, 2006, 7:30:33 PM1/27/06

to SBML Discussion List

frank_bergmann> Does libsbml yet read the biomodels
frank_bergmann> annotation, I think this would be useful
frank_bergmann> to have.

LibSBML does not currently provide special support for
BioModels DB (or any other) annotations. The general goal
in that respect is to provide an API to work with RDF inside
annotations; this would allow other annotations to be read

too. RDF support is definitely on the short-list of
features to implement.

MH

Nicolas Le Novère

unread,

Jan 28, 2006, 6:14:10 AM1/28/06

to SBML Discussion List

> I should also note that I don't think biomoidels.net uses notes
> but uses rdf inside an annotation instead.

To do what? Notes are meant to store xhtml to be displayed to the users.
Metadata is meant to be stored in annotations. This has nothing to do with
BioModels DB. This is SBML Level 2 Version 1.

> Does libsbml yet read the
> biomodels annotation, I think this would be useful to have.

Based on the former votes, the simplified RDF framework will be part of
L2V2, and libSBML will have to support it. Food for the hackathon? We can
help here, because we have the support in BioModels DB and SBMLeditor
(qualifiers are still experimental in the editor though)

--
Nicolas LE NOVÈRE, Computational Neurobiology,
EMBL-EBI, Wellcome-Trust Genome Campus, Hinxton, Cambridge, CB10 1SD, UK
Tel: +44(0)1223 494 521, Fax: +44(0)1223 494 468, Mob: +33(0)689218676
http://www.ebi.ac.uk/~lenov

Herbert Sauro

unread,

Jan 29, 2006, 9:58:25 PM1/29/06

to SBML Discussion List

I am sorry I asked.
Herbert Sauro

Reply all

Reply to author

Forward