Forums

denney> This is what I've been doing, but it doesn't
denney> really seem like the best solution, and I was
denney> thinking this would be a place where the standard
denney> could use some improvement. I didn't know why the
denney> flag was there-- it seems a bit specific (and it
denney> could be detected other ways).

Well, the specification does try to make clear why the flag
is there: e.g., in section 4.9.3 in L2v2, it explains that
the reversibility of a reaction may be evident from the rate
expression, but that the flag has to be there because the
rate expression is *optional*. (If a reaction does not have
a rate expression, then obviously it wouldn't be possible to
use it to infer reversibility :).)

In more qualitative models lacking rate expressions, the
information about reversibility might be available even when
a rate expression is absent, so (we think) the flag has
utility despite that there is some duplication of
information. Some analysis tools may also be able to do
useful work on a model but not be designed to parse rate
expressions.

The spec uses elementary mode analysis as an example, but
presumably there are other applications.

It is critical that the software tool writing out a model
take care to make the 'reversible' flag consistent with the
reaction. This is just a necessary requirement, but
hopefully not an unusual one. There are other places where
bits of a model have to be consistent with each other.

MH