Forums

> Nicolas LeNovere wrote:
>
> Although SBML compartments are physical in the sense that they have a
> size, they are mainly conceptual containers of species. One can define
> overlapping compartments without problem. This has been used several
times
> to model EGFR signalling for instance. In the example I gave, one
could
> model the signalling reactions in one compartment and the metabolic
> network in another, even if the physical location is the same in the
cell.
> The advantage is that one can modify the volume of the signalling
> compartment without touching the metabolic one (for instance, to embed
> the post-synaptic signalling compartment into a dendritic spine.)

In your example, you considered independently changing the volume of the
"post-synaptic signaling" compartment to embed it into a dendritic spine
(which is a geometrically distinct physical location). I assume that
most of the metabolic pathways are associated with the mitochondria
(another distinct physical location). The interior (cytosol) of the
dendrite would likely be another distinct physical compartment.

I really think we need a different method of grouping pathways/submodels
(maybe we should call them "groups" or "pathways" or "modules"). One
reason is that a functional pathway may span multiple physical
compartments, each with their own "size". For example, CellML has two
categories of groupings, one that is functional and one that is
structural.

Jim.

Jim Schaff
Center for Cell Analysis and Modeling, (vcell.org)
University of Connecticut Heath Center
263 Farmington Ave, MC-1507
Farmington, CT 06030