Question about the stochastic simulation (Gibson+Bruck) of the brusselator model

[Konstantia Georgouli]

Hi,

I am trying to run the stochastic simulation for the  brusselator model (File -> Examples->SBML files..) and I would like to ask you a question regarding the results (time and reaction event fluxes report). Here are the steps that I followed:

First of all, I changed the compartment size from 1 to 1E-18 and also the dropdown menu from compartment to Particle numbers (in order to have small particle numbers 301.107 -1806.64).

Second, I set time course for 100s duration and interval size 1s using Stochastic (Gibson +Bruck) Method (use random seed , random seed =123). From the output assistant, I selected the report (Time and Reaction Event Fluxes). Then, I selected the file for storing the report by pushing the Report button.

After running this time course, the results that I am taking for the time=0 are:

Time (R1).ParticleFlux (R2).ParticleFlux (R3).ParticleFlux (R4).ParticleFlux

0 301.107 16269.4 5420.99 1807

The results for the particleFlux of the R1 and R4 seem okay. The problem is with those of R2 and R3. Based on the formulas in the SBML file, (R2).ParticleFlux  and (R3).ParticleFlux have to be ~5896.78 and  ~3263.95 respectively and not 16269.4 and 5420.99. Is this correct or is there something with the reaction rate evaluation that I have not understood?

I am looking forward to hearing from you.

Thanks,

Konstantia

[Hoops, Stefan (sh9cq)]

Hello Konstantia,

In a stochastic process the values will always differ from one run to
the next unless you fix the seed. The effect of a seed is very much
implementation dependent, i.e., using another software or algorithm
will not lead to the same result. Personally, I have never used a seed
in COPASI unless I have to debug the algorithm itself.

Thanks,
Stefan

--
Stefan Hoops, Ph.D.
Research Associate Professor
Biocomplexity Institute & Initiative
University of Virginia
995 Research Park Boulevard
Charlottesville, VA 22911

Phone: +1 540 570 1301
Email: sho...@virginia.edu

[Konstantia Georgouli]

Dear Stefan,

Thank you very much for your answer. I am sorry if I have confused you with my email, I simply wanted to send you all the conditions that I used for running the time course .

I am trying to interpret the stochastic simulation results for the brusselator model and I cannot understand the results for the R2  and R3 reactions for time =0 in the time and reaction event fluxes report ( the results are the same even if I do not use random seed). Specifically,  I have made the calculations based on the reaction rate formulas and the results do not match with the results received in CopasiUI.

Received:

Time (R1).ParticleFlux (R2).ParticleFlux (R3).ParticleFlux (R4).ParticleFlux

0 301.107 16269.4 5420.99 1807

Expected:

0 301.107 5896.78(1806.64*1806.64*1806.64*k1(1e-6)) 3263.95(1806.64*1806.64*k1(0.001)) 1807

Is it something that I have not understood regarding these calculations? 

I am looking forward to hearing from you.

Thank you.

Kind regards,

Konstantia

[Hoops, Stefan (sh9cq)]

Hello Konstantia,

Sorry that I misunderstood your question. The reaction fluxes are not
calculate based on the particle numbers. They are based on the
concentration in your case that would be:
3.0^3 mmol/s = 27 mmol/s

This mmol will than be converted in # using Avogadro's #: 6.0221415e23
and the compartment volume 1e-18.

Which gives you 16269.4

Thanks,
Stefan

[Konstantia Georgouli]

Dear Stefan,

Thank you very much for your help.

Kind regards,

Konstantia