Correct use of "intensive" vs "extensive properties" in SBML specs

[Paul Lang]

In the SBML specs on page 78 I find the following statement:

In multicompartmental models, to be able to specify a rate law only once and then use it unmodified in equations for different species, the rate law needs to be expressed in terms of an intensive property, that is, species quantity/time, rather than the extensive property typically used, species quantity/size/time.

However, the Wiki article states for example that concentration is an intensive property and amount is an extensive property. So I am a bit confused because I believe it should be:

In multicompartmental models, to be able to specify a rate law only once and then use it unmodified in equations for different species, the rate law needs to be expressed in terms of an extensive property, that is, species quantity/time, rather than the intensive property typically used, species quantity/size/time.

What are your thoughts?

Thanks,
Paul

[Hoops, Stefan (sh9cq)]

Hello Paul,

You are correct! The units are correct but the classification as
intensive or extensive is wrong.

An intensive property is a property which does not change with the size
of the system. Examples for intensive properties can be found:
https://en.wikipedia.org/wiki/Intensive_and_extensive_properties#Intensive_properties

Thus the concentration (density) of a species is the intensive property
and the amount is the extensive property. 

What is preserved in any closed system is the extensive property. Thus
for all reactions we should specify the rate law in the extensive
property. If we are dealing with one non changing compartment we can
get away with the intensive property but for multi-compartment or
changing compartments we must use the extensive rate law.

Thanks,
Stefan

--
Stefan Hoops, Ph.D.
Research Associate Professor
Biocomplexity Institute & Initiative
University of Virginia
995 Research Park Boulevard
Charlottesville, VA 22911

Phone: +1 540 570 1301
Email: sho...@virginia.edu

[Paul Lang]

Thanks for the clarification Stefan. Looks like that needs to be updated in the docs then.
Paul

[Lucian Smith]

Thanks for pointing this out!  I've filed an issue report so we don't forget it:

https://sourceforge.net/p/sbml/sbml-specifications/370/

-Lucian

--
You received this message because you are subscribed to the Google Groups "sbml-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbml-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sbml-discuss/3e66a2a1-8d80-46cc-84f3-c493ac6a322dn%40googlegroups.com.

[Michael Hucka]

Hi,

Yes, it does seem like the terms "extensive" and "intensive" have been
flipped. That's unfortunate. Thankfully, those labels are functionally
irrelevant in SBML, because the SBML specifications are careful to
define things in terms of the intended mathematical relationships and
the units involved, and nothing actually depends on the _definitions_ of
extensive and intensive per se. But the documents should still be
corrected -- Paul, thanks for catching this!

As an aside, I wouldn't necessarily take Wikipedia as being the
authority on these matters :-). However, as it turns out, it _does_
match an authoritative and recently-published paper by Jannie Hofmyer on
kinetic modeling of compartmental reaction networks:

https://www.sciencedirect.com/science/article/pii/S0303264720300976

(The paper was recently published, but many people on this list will
recall he's been working on that paper for at least a decade!)

Historical note: I checked the past SBML specs. The use of the terms
"intensive" and "extensive" does not appear in the Level 1 or Level 2
specs, and makes it first appears in SBML Level 3 Version 1 Core Release
1. I speculate that this was just a misunderstanding. I must admit I
still find the definition of intensive and extensive in this context
counterintuitive: the quantity we use that does _not_ involve
concentration is the one that is said to be extensive, i.e., "depends on
the system size". But concentration depends on system size, doesn't it?
So this seems ... backwards.

MH

> --
> You received this message because you are subscribed to the Google
> Groups "sbml-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sbml-discuss...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/sbml-discuss/1be0cc64e1c6b2e348ba72c1fa4d386bcddcd241.camel%40virginia.edu.