Statistical models workshop 2011

20–22 June 2011

The purpose of this meeting is to work on defining a package (or several packages) for supporting statistical models in SBML Level 3. Those include models where the variables are not single values but ranges, distributions or lists of values. Such models are used for instance in population pharmacokinetics.

Some pre-existing work is available about the distrib package.

Programme

Times are UK Summer times.

The audiovisual recordings are done using the EVO teleconferencing system; see the EVO information page for help on playing the recordings. Also available are audio-only recordings in mp3 format—these usually have higher quality than the audio recorded as part of the EVO session.

Monday 20th June

Tuesday 21st June

Wednesday 22th June

Pre and post meeting Background Information

• Everyone should have read the basic introduction to SBML and the detailed summary of the language. But at least a quick look at the SBML specification would be useful.

• A list of reading resources is available in the review by Peter Bonate:Recommended Reading in Population Pharmacokinetic Pharmacodynamics.

• You may want to browse a fairly exhaustive list of Pharmacokinetic Software.

• Pre-existing SBML work: distrib package. This proposal has been discussed during the the 2010 hackathon. Darren Wilkinson has published an account of the discussions.

• Proposal by CellML

• For the harcore, we have a description of the SBML development process.

• More information is available about UncertML. See also JStat.

• A simple example of PK model by Marc Lavielle.

Participants

Some participants attended physically; others attended remotely. The following table lists the attendees.

Funding Acknowledgements

We gratefully acknowledge funding for this workshop from the following agencies:

EMBL-EBI

NIH/NIGMS