Forums

Dear all,

I have a question regarding the notion of the uniqueness of a metaID in an SBML document: Is the metaID necessarily unique in an SBML document and all its versions? Or is the uniqueness restricted to a particular SBML document (i.e. a version of an SBML document)?

Looking into different model versions, we were wondering whether something like the following example is allowed to happen:

model m1.0
<listOfSpecies>
<species metaid="1" name="a" />
<species metaid="2" name="b" />
</listOfSpecies>
<listOfReactions />

... some changes on the model lead to the conclusion that actually species a is not needed for the model, so it is deleted:

model m1.1
<listOfSpecies>
<species metaid="2" name="a" />
<listOfSpecies>
<listOfReactions />

... some more evaluation suddenly detects that a reaction is missing in the model to work properly, so it is added:

model m1.2
<listOfSpecies>
<species metaid="2" name="a" />
<listOfSpecies>
<listOfReactions>
<reaction metaid="1" name="c" />
<listOfReactions />

The question is: Is there something in the SBML spec saying that the newly created reaction in version 1.2 cannot carry an ID that had been existing in a previous version of the model?

If so, that would make keeping track of version changes far easier, so we would be happy if it was the case... although I am not sure how one would be able to control it. Nor who should do it - probably "it is a tool problem"...

The only bit I found in the spec is:
In SBML, type ID is the data type of the metaid attribute on SBase, described in Section 3.2. An important aspect of ID is the XML requirement that a given value of ID must be unique throughout an XML document.

This does however not mention model versions.

Thanks in advance for your comments,
Dagmar

____________________________________________________________
To manage your sbml-discuss list subscription, visit
https://utils.its.caltech.edu/mailman/listinfo/sbml-discuss

For a web interface to the sbml-discuss mailing list, visit
http://sbml.org/Forums/

For questions or feedback about the sbml-discuss list,
contact sbml-team@caltech.edu