Systems Biology Markup Language - Level 1

Mark Addinall mark.addinall at bigpond.com
Mon Apr 30 15:26:14 EST 2001


Herbert Sauro wrote:
> 
> We would like to announce the release of the final specification for Level 1
> of SBML (the Systems Biology Markup Language). It is available from
> 
> http://www.cds.caltech.edu/erato
> 
> SBML is a description language for simulations in systems biology. It is
> oriented towards representing biochemical networks that are common in
> research on a number of topics, including cell signaling pathways, metabolic
> pathways, biochemical reactions, gene regulation, and many others. SBML is
> the product of close collaboration between the teams developing BioSpice
> (http://gobi.lbl.gov/~aparkin/), Gepasi (http://www.gepasi.org/), DBSolve
> (http://websites.ntl.com/~igor.goryanin/), E-Cell http://www.e-cell.org/),
> Jarnac (http://members.tripod.co.uk/sauro/Jarnac.htm), StochSim
> (http://www.zoo.cam.ac.uk/comp-cell/StochSim.html), and Virtual Cell
> (http://www.nrcam.uchc.edu/).
> 
> The motivations for developing SBML stem from the current inability to
> exchange models between biochemical network simulation/analysis tools. The
> lack of a common standard for exchanging models poses two problems for
> researchers.
> 
> First, models cannot be shared directly between tools because the file
> format used by each tool is unique and nontransferable. This makes it
> difficult if not impossible for a researcher to exchange models. Second,
> when a simulator is upgraded or is no longer supported, models developed in
> the `old' system become stranded and unusable in any other simulator. This
> later scenario has already happened on a number of occasions, with the
> resulting loss of the model to the scientific community. With the recent
> roliferation of new biochemical simulators, this situation can only get
> worse.
> 
> We have kept the base definition of SBML as simple as possible, so that
> simulator developers will not find it too difficult to implement support for
> SBML in their tools.
> 
> This base definition, called SBML Level 1, is the result of merging
> modeling-language features from a number of existing software packages, and
> encompasses the minimal information required for biochemical models.
> Additional structures and facilities will be added to SBML in subsequent
> levels; among the features we are developing for SBML Level 2 are
> hierarchical models and spatial characteristics. By freezing subsets of
> features in SBML definitions at incremental levels, we hope to provide the
> community with stable standards to which software authors can design to,
> while at the same time allowing the simulation community to gain experience
> with the language definitions before introducing new elements.
> 
> The effort perhaps closest in spirit to SBML is CellML
> 
> (http://www.cellml.org). CellML is an XML-based markup language designed for
> storing and exchanging computer-based biological models. The constructs in
> CellML tend to be at a more abstract and general level than those in SBML
> Level 1, and describe the structure and underlying mathematics of cellular
> models in a very general way. By contrast, SBML is closer to the internal
> object model used in model analysissoftware. Because SBML Level 1 is being
> developed in the context of interacting with a number of existing simulation
> packages, it is a more concrete language than CellML and may be better
> suited to its purpose of enabling interoperability with existing simulation
> tools. However, CellML offers viable alternative ideas, and the developers
> of SBML and CellML are actively engaged in ensuring that the two
> representations can be translated between each other.

With thanks, could you or I cross post this to aus.science?
Very interesting indeed.

Regards,
Mark Addinall




More information about the Comp-bio mailing list