Re: [Ecell-users] set up ODE for species directly.

[Koichi T. <sh...@e-...>]

Hi Lu,


1. Expression string in an ExpressionFluxProcess represents a flux.

Therefore, your expression means a zero-th order flux.

 > da/dt = -K^2   and  db/dt = K^2

2. In E-Cell, It is the user's responsibility to ensure unit
    consistency.

3. This is a biochemistry question, rather than a software question.
    Your expression is Kch [C1]^2 / (K1 + [C1]^2) ( [C2] - [C1] ).
    It seems like to have a form of a Hill equation multiplied by
    a concentration gradient.   If Kch really has the unit [1/s], there
    must be an additional factor that converts the unit of the second
    term (concentration difference) to the same [1/s].  How you do this
    depends on what you want to model.   If you assume it is
    linearly dependent on the concentration difference,
    you can replace ([C2]-[C1]) with something like lambda ([C2]-[C1]).
    Lambda must have a unit [1/M].

4. Expression processes are continuous processes, therefore
    a continuous Stepper is necessary to integrate them.
    We recommend ODEStepper for practically all types of ODE models.

Koichi




> Hi Moriyoshi and Yuri
> 
> Thank you very much, yes I think ExpressionFluxProcess will be helpful.
> But I have several questions:
> 
> 1. ExpressionFluxProcess represents a customized reaction rate or it
> directly presents flux?
> e.g.
> Process ExpressionFluxProcess(a2b)
> {
> k 33;
> Expression "K^2";
> VariableReferenceList [S0 Variable:/cell:a -1][P0 Variable:/cell:b 1]
> }
> 
> Then, if we only consider this one reaction, which of the following
> expression is correct?
> da/dt = -K^2[a]   and    db/dt = K^2[a]
> or
> da/dt = -K^2   and  db/dt = K^2
> 
> 2. Since the unit for Expression must be "number/second" instead of
> "concentration/second", if, in the expression, the MolarConc of some
> variables are used, I must multiply self.getSuperSystem().sizeN_A once, in
> order to inform the system that I changed the unit?
> I just wonder how e-cell do this consistency check ?
> If under some circumstances, I need to multiply the size twice.....can
> e-cell detect it?
> 
> 3 In the case of multi-compartment model, in which way I can indicate that
> I multiplied correct compartment/compartments?
> e.g.
> Process ExpressionFluxProcess(Ca_ER2Cytosol)
> {
> Kch 3;
> K1 2;
> Expression "(Kch*C1.MolarConc^2/(K1^2+C1.MolarConc^2)) *
> (C2.MolarConc-C1.MolarConc)";   #this expression is wrong since I haven't
> multiply any compartment size!#
> VariableReferenceList [C1 Variable:/cytosol:Ca 1][C2 Variable:/ER:Ca -1];
> }
> 
> * in this example I want to represent ER release Ca to cytosol.
> the first part of this expression,
> Kch*C1.MolarConc^2/(K1^2+C1.MolarConc^2) should have no unit problem,
> since the unit for Kch is per-second, and the unit for K1 is
> concentration.
> However, for the second part C2.MolarConc-C1.MolarConc, I should convert
> its unit into Number instead of concentration. Which compartment size, and
> what kind of expression I should use here.
> 
> 4 which stepper I should use for ExpressioFluxProcess?
> should it must be a continuous stepper?
> Can I use ODEstepper?
> 
> 
> I am looking forward to your reply.
> Thank you very much !
> 
> yours Lu
> 
> 
>> Hi,
>>
>> In which sence did you say "directly" ? Sounds like
>> it's ExpressionFluxProcess that fits to your case; did
>> you try it yet?
>>
>> Moriyoshi
>>
>>
>> 2008/3/18, Lu Li <lu...@eb...>:
>>> Hello
>>>
>>>  I wonder in e-cell, can I directly set up ODE for some species?
>>>  And these species, even can play a role in some reactions, their flux
>>>  will not be influenced by these reactions?
>>>
>>>  I am looking forward to your reply.
>>>
>>>  Best Wishes
>>>  yours Lu Li
>>>  --
>>>  Lu Li
>>>  PhD student in Computational Neurobiology
>>>  EMBL-European Bioinformatics Institute, Cambridge, UK
>>>
>>>  -------------------------------------------------------------------------
>>>  This SF.net email is sponsored by: Microsoft
>>>  Defy all challenges. Microsoft(R) Visual Studio 2008.
>>>  http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>>>  _______________________________________________
>>>  Ecell-users mailing list
>>>  Ece...@li...
>>>  https://lists.sourceforge.net/lists/listinfo/ecell-users
>>>
>>
>> --
>> Moriyoshi Koizumi <mor...@gm...>
>> (also reachable on <mo...@sf...>)
>>
> 
>