Redundant information in model structure about cobratoolbox HOT 4 CLOSED

Redundant information I do not view as a problem in the model structure itself. Correct me if I'm wrong but I think grxnGeneMat and rules are actually generated from grRules when a model is imported from sbml.

rev is indeed another long story

from cobratoolbox.

Actually, I really think that it is a problem. Aside from the fact that not everyone uses SBML, from the perspective of a Matlab user this is inconvenient. Every time one wants to change something on the structure, one has to keep in mind that there are several elements that contain that same information.

from cobratoolbox.

It is a bit of an issue, but I am not sure the alternative is better. The
rxnGeneMat is used frequently during exploratory analyses and probably by a
lot of functions in the COBRA toolbox. so if is gotten rid of, a user would
have to generate it every time they want to use it. Furthermore, new users
may not know about it as easily. My impression, which could be wrong, is
that the gprules are not frequently changed by most COBRA users.

I'm not totally against the removals but want to point out there are some
potential side effects that may outweigh the benefits.

This may be totally wrong but could a listener be used to allow automatic
updates to the other fields?
http://www.mathworks.com/help/matlab/ref/handle.addlistener.html
I'm trying to find a solution that doesn't involve full blown abstraction
of the model and require getters and setters for all fields.

On Fri, Mar 13, 2015 at 4:49 AM, Alberto Noronha [email protected]
wrote:

Actually, I really think that it is a problem. Aside from the fact that
not everyone uses SBML, from the perspective of a Matlab user this is
inconvenient. Every time one wants to change something on the structure,
one has to keep in mind that there are several elements that contain that
same information.

—
Reply to this email directly or view it on GitHub
#40 (comment)
.

Brandon Barker
[email protected]

from cobratoolbox.

Having a function that generates the matrix would solve this issue, I believe. And that would ensure that the matrix is coherent with the grRules.

from cobratoolbox.