#215 How to interpret constant species in variable-sized compart.

[Michael Hucka]

Raised by Lucian during the SBML Editors' meeting during HARMONY 2012:

Consider a model with a species with no rules about how the species itself changes (no reactions, no rules), with the attribute 'hasOnlySubstanceUnits=false', and which is in a compartment whose size changes. Its amount does not change, but its concentration does. Is it valid to give this species the attribute 'constant=true', or must it be flagged 'constant=false'?

This is never explicitly stated in the spec.

[Lucian Smith]

The conclusion was that, even though the concentration will change, you can still flag this as 'constant=true' so that an algebraic rule can tell whether or not it may or may not determine this value.

[Lucian Smith]

I agree with the proposed change and that it should be done.

[Chris Myers]

I am accepting this issue as valid.

[Nicolas Le Novère]

I am accepting this issue as valid.

[Lucian Smith]

Considering we decided not to change the validation rule for this, but only add a clarification, this can be an L2v4 spec edit, not a l2v5-only change.

[Nicolas Le Novère]

I see no point in a L2V4 spec edit since we are releasing L2V5.

[Michael Hucka]

There is a misunderstanding about the categories of bug reports. A report like this, about the L2v4 specification, is something to be fixed by issuing either a release 2, or an L2v5. The report, however, is still *about* L2v4. That is why it belongs here. We then use the reports as the basis for constructing the changes to be put in L2v5 (or if the change is small enough, L2v4r2).

[Lucian Smith]

We actually discussed and came to a conclusion on this issue at an editor's meeting several years ago. The upshot was that the 'constant' flag on a Species a) references its amount, not (directly) the concentration, and b) is needed for algebraic rules.

See https://sourceforge.net/p/sbml/sbml-specifications/211/ for another aspect of this same issue.

[Michael Hucka]

I accept this issue as valid, and agree with the proposed solution.

[Dagmar Waltemath]

I accept this issue as valid, and agree with the proposed solution.

[Lucian Smith]

Fixed in SVN for both L2v5 and L3v2, and will be part of the forthcoming release of those specifications.