ABI CellML Meeting Minutes, 2nd March 2011

This week's agenda

1. CellML / PMR2 Tutorial: Progress:
   • Randall noted that he has met with the Engineering Science department IT staff and given them the info for setting up the lab computers.
   • Via Randall - Dougal Cowan has put together a tutorial. There are already enough people to help run it, but more would be better.
2. Metadata specifications:
   • Progress: Mike Cooling has presented a draft metadata specification; it consists of core and satellite specifications.
   • Discussion:
     • Peter asked about the link between models and ontologies, and whether Mike was aware of Sarala Wimalaratne's work for the RICORDO project on a web-accessible metadata query and management system.
     • There was extensive discussion about where metadata should reside (on the model vs in a centralised database). Consensus was generally that metadata linking a component to a biological entity is best in the model, but such data is only added with the authorisation of the author. Otherwise, the person adding the annotation can create a fork of the model, or store the annotations externally.
     • Another issue discussed was the inheritance of metadata when a model is modified if the model and the metadata are not co-located, i.e. the metadata about one version of a model in PMR2 may still apply to the next version of that model, but this link would not be made if the metadata is located in the Ricordo system and the Ricordo system refers to models in PMR2 by their URL.
     • Action item 1: Randall will put Mike and Sarala in touch on this issue.
3. Physiome Model Repository version 2:
   • Progress: Tommy has been working on a clean up of PMR2. He has generalised authentication to support different protocols, moved the Mercurial part to the Mercurial module, and split the code up into modules.
   • Planning: Randall posted a comment to tracker item 1909 which was based on his response to Alan Garny's e-mail query regarding PMR2 web-services. It outlines possible web-services that might be of relevance to OpenCOR/OpenCell.
4. Repository contributions:
   • Planning: Peter noted that Hanne Nielsen is continuing to work on SVG diagrams and session files.
   • Discussion: Andre noted that someone submitted a model and requested it be published via the site, but there is no process on the site to publish it; it only got published after the author asked Andre, and Andre asked Tommy and Dougal Cowan. At an upcoming retreat, Peter will ask for curators for each group at the ABI. Until then, Dougal can follow the RSS feed to identify things to publish. Action item 2: Tommy will check that there is an RSS feed, and add one if necessary, and then Randall will check in with Dougal.
5. Specifications: Progress: There was a discussion at a breakaway meeting last week on data types.
6. CellML API: Progress:
   • Tommy wrote a CellML API patch a few weeks ago to add a path to the list of paths used when building the Python bindings, fixing a compilation error when the CellML API is built for the first time with Python support enabled, and Andrew has now pushed this as Jon Olav Vik was encountering the same problem.
   • Andrew has been continuing work on the SEDML Processing Service (SProS). Most of the implementation is now complete, but after this, a testsuite needs to be written, and significantly, a way to work with the SProS SED-ML is required.
   • Andre reports that Amy's C# based SED-ML simulator is working, at least on her machine.
   • Andre also reports that the API in OpenCMISS is working now.
7. Other business: Andre and Peter noted that ICSB, EMBC, and Research Week were all on in the week beginning August 29th; we need to decide if we want to do a tutorial at any of them fairly soon as deadlines are approaching.