Constructing a large model with few parameters

[at...@gmx.de]

Hi all.

First of all: The btk-mca newsgroups - the one on yahoogroups.com and
the one on bio.net - have been very inspiring. It is very unfortunate
that open discussions like these are hard to find. If anyone can hint me
to more material on the web with this quality, please write.

On topic:
I would like to build a bacterial model with ca. 200 reactions and as
few parameters as possible. I'm not interested in exact reproduction of
flux and concentration data, but the trends (rises & falls) would be
sufficient. Therefore, my first idea is to build a model with mass
action kinetics (which makes use of equilibrium constants and has only
one parameter left) and add only a few more detailed rate laws at
critical points.

Questions:
1. Does this make sense? What would be a better approach?
2. How do I identify these critical points (e.g. large equilibrium
constant, enzymes which are regulated by feedback inhibition, ...)?

Looking forward to your suggestions,

Kai

PS: Cross-posting is intended, please don't get offended.