French Working Group on XML Markup Languages for modeling in physiology and biochemistry

(organized 4 workshop meetings in 2004 and 1 in 2005)

Organizer: Randy Thomas
INSERM U.467, CHU Necker, Paris (now at IBISC CNRS FRE 3190, Evry)
Cette adresse email est protégée contre les robots des spammeurs, vous devez activer Javascript pour la voir.

XML markup languages (ML) are proving useful in a wide variety of fields as an efficient, platform-independent, extensible, open source medium for exchange of documents, models, or data among projects having a common lexical base or a common set of concepts but which operate in a variety of computing environments or are implemented in different programming languages. Two such markup languages have achieved a measure of popularity in the modeling of biochemical networks and cell physiology & electrochemistry, namely CellML (www.cellml.org), developed principally by the group of Peter Hunter (New Zealand) in collaboration with Denis Noble (Oxford and Princeton), and the Systems Biology Markup Language, SBML (www.sbml.org) at Berkeley. The degree of success of these initial efforts can be gauged by the number (more than 30 so far) of simulation languages or ODE solvers (especially those that target biochemical networks and Metabolic Control Analysis) that now use one or both of these MLs as a native input format for model description, and (2) the growing number of legacy models that have been translated into these MLs (e.g., see the list at http://www.cellml.org/examples/repository/index.html) in the interest of creating a common pool of pluggable building blocks to be used in the construction of more integrated or global models.

In France, a small but growing number of laboratories has begun using this approach for development of their models in cell physiology and metabolism, and a second group of laboratories is interested in building on the CellML/SBML groundwork to develop similar markup languages adapted to the modeling not just of cell or metabolic-level problems but also of physiological function at the tissue, organ, or inter-organ levels. Progress, however, has been slow due to the difficulty of identifying and mastering the relevant software tools of this rapidly developing new technology and to the isolation of the interested labs from each other and from the rest of the international community. Added to this is the fact that these labs have varying degrees of expertise in computer science and not all have been able to establish a collaboration with a computer science laboratory.
We propose to create a working group on this subject to bring these laboratories into contact in order to facilitate their cooperation in this new technology. This should be helpful, since a large part of the difficulty arises from the need to choose appropriate generic software tools for the editing, parsing, and processing of the XML files central to the technique and therefore common to all such implementations. Upstream from this is the need to educate in the XML tools, i.e., XSLT, DOM, XPath, etc., and the larger project of defining the schemas (or ontologies) specific to each domain of application, a pre-requisite for development of a markup language. Downstream from acquisition/mastery of the generic software tools is the development of interface programs for the use of ML model-descriptions with in-house numerical methods.
The working group will be an open association of laboratories with this common interest but whose individual research domains vary considerably. It is anticipated that the working group will
- meet periodically for one- or two-day working sessions to:
o share experiences with existing MLs
o review and compare existing software tools, including not only those used for creating and managing MLs but also computing environments (e.g., ode simulators) accepting SBML or CellML model descriptions (of which there are now more than 30 listed on the SBML website)
o organize the development of extensions to CellML and SBML
- work together on specific aspects electronically during the interim periods
- establish collaborations/consultations with STIC labs having appropriate computer science expertise
- establish contact with international laboratories working in this field
- establish a website for presentation and distribution of software tools and reports of progress
- organize a Journée Scientifique in the GdR STIC-Santé,
- prepare a report, for the GdR, on the state of the art of these XML markup languages, and
- if enough interest in this technology is manifested it may become appropriate, to propose a Training Workshop (Atelier de Formation).
Organization: A Principal Organizer (or Convener) (R Thomas) and a Core Scientific Committee (the list below)