ABI CellML Meeting Minutes, 2nd February 2011

Present: Dougal Cowan, Andrew Miller, Amy You, Randall Britten, Tommy Yu, Poul Nielsen, David "Andre" Nickerson

Last week's action items:

1) Tommy - check that there is proper disaster recovery (usable backup) for PMR2.

  • Done - this is in place. A disaster recovery simulation needs to be run to test that this is all working properly.

2) Tommy - remove global workspaces list link from main navigation bar.

  • Done - also made user workspaces the default.


1) CellML API - update


  • Andrew: working on the SED-ML processing service.


  • Andre: why don't we use libsbml to process SED-ML?
  • There was a discussion about what the SED-ML processing service will do, and why it makes sense to create it.
  • Andrew: it may make sense to provide the SED-ML processing modules separately from the API. SED-ML support is being added to enable descriptions of stochastic modelling.
  • There was a discussion of stochastic modelling, and whether parameter variation descriptions should appear in the CellML or SED-ML.
  • Poul: is there a mechanism in SED-ML for specifying pseudo-random (seeded) numbers? Andrew: no. Andre: there probably won't be, as this is being added to SBML.
  • Poul: we need to push to get stochastic modelling support included in SED-ML.
  • There was a discussion about SED-ML implementations, and the languages they are available from.
  • Andre: has re-opened tracker item 2085 about excessive warnings. Also getting random crashes with the Visual Studio linker.
  • Andrew: are the binaries working? Andre: on Andre's 32 bit machine they are working, but not on Amy's 64bit Windows.
  • Andre: we need to be able to go from xpath in SED-ML to a unique identifier to use in runcellml.
  • There was an extensive discussion about this problem, and what to do about xpath vs imports in CellML.
  • There was a discussion about the differences between imports and xincludes in CellML and SBML.

2) CellML core specs - update


  • Andrew is writing up his ModML work.

3) Repository contributions - update


  • Mike Cooling will be looking at Mona's models.

4) PMR2 development - update


  • Tommy: finished making the protocol generic, moved out the Mercurial side so that workspaces can have a generic backend.
  • Randall: do we want to implement mixed technologies? Probably not.
  • Tommy: the new separation is cleaner and makes testing easier.
  • Tommy: working on the MathML view. This is not yet 1.1 compatible.
  • There was a discussion about PMR2 version numbers.

5) ABI summer tutorials:CellML tutorial and PMR2 tutorial


  • Randall: can Dougal bring the PKPD tutorial along to the next meeting to see if it is going to be useful for the CellML tutorial.
  • Randall: could the PMR2 tutorial be run on a temporarily staged repository or a VM, so that people can make exposures in the tutorial?
  • There was a discussion about how to provide PMR2 access, either by a VM or a sandbox server which is periodically reset.

6) PKPD tutorial for PAGANZ


  • Poul: we will need some volunteer tutors, and to do a run through.
  • Randall: happy to be a tutor.