Meeting Minutes 19 December 2007

Present: Randall Britten, Sarala Dissanayake, Peter Hunter, James Lawson, Catherine Lloyd, Justin Marsh, Andrew Miller, Poul Nielsen, Tommy Yu

Sarala's presentation on her work on generating a graphical representation of a biological model

  • CellML model – generating visual diagram to show the underlying biological model.

    • CellML -> CellMLOWL -> CellMLBioOWL -> SVG Visual

      • In the scope of CellMLOwl, each element has an ontology, which is what Sarala generates.

  • CellML imports normally don't have access to variables inside the encapsulated components.

  • RDF is then used to append metadata in the CellMLOWL domain with the range in the CellML components to enable addressing the objects inside the imported components.

  • The cmeta:id of the variable is annotated with RDF (an RDF graph about the cmeta:id), however the problem is, if the model was imported twice, the flattened model will have duplicate cmeta:id

    • Poul pointed out this is not a concern, as each individual imports should have its own identifier, which are unique.

  • Sarala also introduced methods to collapse groups of links such as entity <–> process <–> entity into entity to aid diagram generation, also describes how she split up the model based on the biophysical level and the biological level.

    • There were further discussions on the specifics of collapsing groups of entities and process into an entity or process.

Review of last week's action items

  • Tommy will implement this new model download page

    • He has not start working on it yet.

    • Peter pointed out that the design should also include pointing out that the PCEnv session link is preferred if it exists.

    • This item will now be tracked by the tracker.

  • Poul to look for those minutes (pre-2001) (on the naming convention)

    • Poul has found something but not relevant, but gave what he found to Andrew anyway.

    • The decision to allow an identifier that begins in underscore(s) with a numeric digit following right after will need to be re-evaluated with the next specification.

    • Action item: Andrew will check the tracker for the issue of variable naming.

  • James to change the current development page to include search links to the tracker with relevant explanations.

    • James will remove the outdated page and will add links to the tracker.

  • Andrew needs to come up with terminologies for different model types

    • See below.


CellML 1.2

  • Andrew did sent out the email soliciting suggestions on which terminologies to define that will aid the explanation of the difference between between models at different stages of imports.

    • There has not been much feedback.

    • CellML XML Infoset may possibly be a candidate term. Not quite resolved yet.

  • Normative specification is nearly done. Andrew has rewritten it in a more technical style of writing.

  • Perhaps developing the informative document in parallel might assist in finding discrepancy between these two documents sooner. Involving more people in this process will help also.

  • The time frame to finish this finish these documents is uncertain, as Andrew is doing the bulk of the writing and has other responsibilities. Justin could possibly help.

PCEnv Renaming, Prominence on the CellML website

  • Alan wouldn't be too concerned about the timing: we have known for months that we want to rename PCEnv/COR, so... If anything, he suggests we probably want to come up with a new name by March, i.e. by the time of the next CellML workshop.
  • However it was noted the purpose of PCEnv may be redefined at that time, thus it's better to wait until then before deciding for a suitable name.

  • As for the giving PCEnv special attention on the CellML website, Andre does not want to give one particular tool too much emphasis. So making a quick links section to point to PCEnv is also deferred.

  • Despite of that, there needs to be more clear instructions for new users of CellML to know where to get the tools, for instance when they are browsing models in the repository, they need to know how to run the models there.

CellML workshop/meeting 2008

  • The rough outline for the programme of the workshop is put together. Twelve speakers were identified.

  • The Japanese group will be sending three or four people, but Taishin Nomura and Yoshihisa Kurachi will not be attending the conference as there is a different conference in Japan they will be attending. However, they will be visiting Auckland on 21 and 22 February instead to do discussions on ontologies in particular.


  • Tommy still picking up specifics of Zope 3 and Plone 3, but he has put what he originally did in Plone 2.5 partially into the Plone 3 framework.

Final Notes
  • This is the last meeting of this year. Next meeting will be held 9 January 2008.