> Hello Fede,
>
> You can simply add a reaction A{compartment_1} -> A{compartment_2} in COPASI. If you use mass action kinetics this will result in the same ODEs as Fick's law. What Juergen is talking about is that the k needs to be properly scaled with the volume. COPASI does scaling automatically for inner compartment reactions.
>
> Thanks,
> Stefan
Hi Stefan,
I'm actually having a similar issue and I wanted to clarify what you mean by 'k needs to be properly scaled by volume'. Say for example I'm moving molecule A into the nucleus (A -> nA). Mass action kinetics means that the rate would be k*A. By scale correctly, do you mean change this law to k*Nuclear_Volume*A? ... Further, if I were moving molecule A out of the nucleus it would be k*Cytoplasmic_Volume*A?
Thanks - your help is much appreciated
> Hi Stefan,
> I'm actually having a similar issue and I wanted to clarify what you mean by 'k needs to be properly scaled by volume'. Say for example I'm moving molecule A into the nucleus (A -> nA). Mass action kinetics means that the rate would be k*A. By scale correctly, do you mean change this law to k*Nuclear_Volume*A? ... Further, if I were moving molecule A out of the nucleus it would be k*Cytoplasmic_Volume*A?
Hello Ciaran,
For a reaction involving multiple compartments the kinetic law needs to specify change of the particle numbers per time. Since the rate law still depends an the concentrations of the species in the individual compartments you need will have to adjust the kinetic constant by multiplying with the quantity to number factor and a scaling factor. Normally this scaling factor would be the size of the compartment in which the reaction occurs, which would be in your case the membrane enclosing the nucleus.
Thanks,
Stefan
Aaahh right, my confusion was that I didn't know which scaling factor to multiply my rate constant by (i.e. the starting point or the destination). I haven't actually included a nuclear membrane in my model as I was tryin to keep it simple, I'll add one and see if I have more luck. Cheers