Forums

[Let's assume there was a misunderstanding earlier between a discussion on
the rate-law (SBML business) and the ODEs (simulator business)]

> However that implies (at least for the reactions that take place in one
> compartment) that the k value implicitly contains the compartment volume
> since usually you assume that the reaction scales with compartment volume
> (i.e. if the volume doubles the amount-of-substance-rate of a first order
> reaction should also double).

It is not just you assume. It is the physical basis of chemical kinetics.
I think...

> So far it would be (quite) ok, just not nice. We should still not
> encourage
> people to write the models like this.

No, we shouldn't. In addition, you cannot apply this with non-linear
rate-laws such as Henri-Michaelis-Menten etc. I actually found ugly things
such as (Vmax * species) / (Km + species)
With species having initialAmount and hasOnlySubstanceUnits="true". The
authors were happy because the rate-law provided substance/time!!!

As stated before in another thread, if one uses continuous chemical
kinetics in aqueous solution, one should always use the concentrations as
activities.
And then multiply that rate-law by the adequate volume to get the SBML
kineticLaw.

--
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