ABI CellML Meeting Minutes, 14th April 2010

Present: Randall Britten, Justin Marsh, Tommy Yu, Geoff Nunns, Catherine Lloyd, David "Andre" Nickerson, Andrew Miller, Poul Nielsen (joined later)

Last week's action items:


This week's agenda:

  1. E-notation: Catherine requested that a script be run through the repository to correct all the broken e-notation.
    Tommy says that he plans to do this after the beta release of PMR2. This will probably be done next month.
  2. Andre asked about tracker item 2504 about the XML Schema Definition.
    It seems to have fallen somewhat between the cracks, because no one got CCd to it automatically. It was decided that Andrew and Justin should, at least, be the default CC for CellML Specification issues.
    Andrew asked whether we should maintain the XSD now that we have a validation service and want to encourage people to use that instead, since it does everything the XSD will be able to do and more in terms of validation. Justin and Andre noted that XSDs are useful for more than validation. Andrew is not sure that it is possible to fix the issue, but Brandon claims to have a fix.
    Action item 1: Justin will post a comment on the tracker item asking for the fixes to the XSD to be attached to the tracker item.
  3. Andre noted that the tracker item filed by Lucian on curation had no CCs except James.
    Action item 2: Andrew will go through all tracker components and identify ones which are missing default CCs. Once Andrew is done, Randall will look into who needs to be the default CC on any remaining items.
  4. Catherine has been in touch with Peter Robinson, who asked if any progress has been made on delayed differential equations.
    Randall noted that Mark Hanners will be working on delayed differential equations in CellML as part of a fourth year project.
  5. PMR2 development - update
    Tommy has been working on the bulk exposure editor. He can define profiles for an exposure file, and can have views created.
    There was discussion of how the testing by users that James used to do should be replaced.
  6. Repository contributions - update
    Catherine has been working on the Bertram pancreatic bursting model. This was one of the very few of Bertram's models which did not have code on his site, but he has sent Catherine the code. Geoff noted that the reason it wasn't working could be because it has reset rules - but he is not sure, as it appears that the variables being reset by the reset rules are not used by the rest of the model.
    Geoff has finished the Vinnakota 2006, Vinnakota 2010, and Albert 2005 glycolysis models.
    Catherine noted that Lucian had asked about some models in the repository which had varying extensions other than .cellml. In particular, some models had .CellML extensions. Andrew noted that the official recommended extension according to RFC4708 is .cellml, and using anything else is not best practice. Tommy noted that it makes no practical difference to the operation of the repository. In addition, Catherine noted that some models contained a \ rather than / as a path separator. Consensus was that this was wrong, and that future versions of OpenCell should be able to produce a validation warning about this.
  7. CellML API - update
    Andrew has developed a Java wrapper and sent it to Morgan and Allyson for them to use with Saint. At present, Morgan reports that it crashes the JVM. Andrew has started looking into this. Morgan also requested an additional method; Andrew plans to look into whether what they want can be done with the existing wrapper, and get back to Morgan and Allyson.
    He has also changed the CellML Integration Service to use the latest version of SUNDIALS, rather than an older stand-alone version of CVODE, in preparation for finishing IDA support.
  8. OpenCell - update
    Justin has been working on the prototype using Scala. He has changed from using JFace to using Eclipse, and has something working demonstrating tabs. Andrew asked about his experience, and Justin said that there has been a learning curve, but overall, it has been easier to work with than C++ and plain Java. He is not yet able to judge whether it is easier to work with than Javascript / XUL.
  9. CellML Specification - update
    ModML - Andrew continues work on the reaction support library. Randall asked if Andrew has asked to see Dan Beard's MATLAB scripts; Andrew hasn't at this stage, but might consider it.
  10. Metadata specifications - update
    No update this week.
  11. cellml.org - update
    Catherine noted that Dougal has created a template workspace, including a session and SVG figure. Tutorials will also be added to this workspace.
  12. Additional item - papers on PMR2
    Catherine noted that abstracts for ICSB are due by 3rd May. In addition, Jacky Snoep has contacted her to say that the essays for an encyclopedia are due by the end of the month.
    There was extensive discussion about what articles we should be working on with respect to PMR2, where they should be published, and how they should differ from each other to ensure that the same thing isn't submitted for publication twice. After discussion, the following table was put together:
    Article What is covered Authorship
    Encyclopedia entry - Jackie Snoop 3 pages overview of PMR2 – from end-user perspective. Catherine
    ICSB - abstract first Broad overview from user perspective. Andre and Catherine + many others
    Bioinformatics Application Note 2 pages – description of software. List of features Tommy & Catherine
    Article – DVCS for Biological Model Repositories Rationale for DVCS Andrew + many others
    Online-only application note Longer rationale / description for a later version of PMR2 Tommy + many others
    Best practice research article How to build modular models Mike Cooling + many others