ABI CellML Meeting Minutes, 5th May 2010

Present: Randall Britten, Dougal Cowan, Andrew Miller, Mike Cooling, Tommy Yu, Geoff Nunns, Sarala Wimalaratne, Peter Hunter, Poul Nielsen, Justin Marsh

Apologies: Catherine Lloyd, David "Andre" Nickerson

Last week's action items:

1) Catherine to follow up on membership of biocuration.org.

  • Catherine will join in July, when the membership year starts.

This week's agenda:

1) PMR2 development - update

  • Tommy has almost finished writing the application note.
  • Randall said that we need to include authors who contributed ideas to PMR2.
  • There was a discussion about who to include as authors.
  • Peter asked how to get to the CMGUI/FieldML example on the repository.
  • Randall said that the example is only up on the staging instance at the moment. There was a discussion about the FieldML viewer, which will be quite basic in the next release of PMR2 - it will only have rotation of an opaque, white representation of the model.
  • Peter asked when the example (and the next version of PMR2) will be live.
  • Tommy and Randall estimated that there are 2 or so more weeks of development, then a further 2 of staging and testing. It could go live at the end of the month, or a little later.
  • Tommy reported that URL shortening has been completed.
  • There was a discussion of the PMR2 plan; completed and outstanding items.
  • Peter asked if the aliases (models.cellml.org, models.fieldml.org, models.physiomeproject.org) could be implemented. Randall said that the filtering of models needs to be sorted out before these aliases go live.
  • Poul suggested that Catherine be reminded to give Tommy the latest version of the terms of use, so that he can put it up on the site.
  • Andrew has been analysing the changes in number of revisions being submitted to the repository since the DVCS (Mercurial) has been in place. There is a clear correlation between PMR2 and an increase in the number of revisions being submitted.
  • There was a discussion of the analysis presented by Andrew.

2) Repository contributions - update

  • Geoff has added two models, which are not yet exposed. These are Maurya et al 2005 and Bornheimer et al 2004. Geoff has a lot more of these models to add.
  • Geoff has been working through the calcium dynamics section, having finished the metabolism section.
  • Peter asked if there are session files for these new models - Geoff said that there are not, due to technical difficulties in validating the results of the models.
  • Peter asked how Hanne is being assigned sessions for creation - Catherine is going to be assigning her models by email.

3) CellML core specs - update

  • Poul reported that there is some urgency in writing up the specs for infintessimal and finite delays, even if they don't currently work in the API.
  • Randall said that we could generate fixed delay MATLAB code, which might make our validation of the apporach quicker.
  • Randall said that we are close to wrapping up infintessimal delays.
  • There was a discussion about developing specifications for delays.

4) OpenCell development - Qt/C++ versus JVM/Scala, and what to do next?

  • Randall said that support for IDA is important enough to release a new version of the XULRunner OpenCell.
  • Justin set up an EVO meeting with Alan Garny, to show Alan the JVM/Scala prototype.
  • Alan was amenable to the idea of developing a metadata editor in JVM/Scala, and then assess the potential for using this platform for OpenCell.
  • Peter suggested having a meeting on the platform choice issue, including other software development teams from ABI - this is relevant to the CMGUI WX replacement. The meeting will have to wait for Andre to return, and should include Alan Garny by conference call.
  • Mike asked about how the platform change might affect the API, particularly with respect to Java support.
  • Randall said that JVM/Scala development would push support for Java.
  • Sarala asked if she could have some assistance getting the CellML API working on her Windows 7 64bit machine.
  • Peter said that a summary of the Qt/C++ versus JVM/Scala analysis would be useful for other development groups in ABI.

5) Metadata specifications (general, citations, graphing, simulation, other) - update

  • Randall said that we should wait for a report from Andre.

6) CellML API - update

  • Randall said that it would be very useful to have support for initial conditions metadata, for Geoff's work.

7) Time permitting: Updates or additional items from volunteering attendees

  • Sarala reported that she is working on a framework to allow searching across multiple databases based on metadata. She requires a metadata spec and API for CellML.
  • Sarala described how a grammar is required in order to form useful composite searches from terms such as "blood pressure" where each individual term would be from a different ontology.
  • Sarala's goal is to be able to search across a number of databases using MIRIAM annotations.
  • Peter asked about how clinical databases are annotated.
  • There was a discussion about clinical annotations.
  • Peter asked if the MIRIAM standard will be appropriate for annotating FieldML models. Poul could see no reason why CellML and FieldML could not use the same annotation scheme. There followed a discussion about annotation of FieldML and spatial models in general.
  • Peter said that a high-level description of the aims of the metadata editor should be developed before the project is started.