Team:Grenoble/Modeling/Amplification/Quorum

From 2012.igem.org

(Difference between revisions)
Line 40: Line 40:
<center><img src="https://static.igem.org/mediawiki/2012/4/48/Eq36_grenoble.png" alt="" /></center>
<center><img src="https://static.igem.org/mediawiki/2012/4/48/Eq36_grenoble.png" alt="" /></center>
</br>
</br>
 +
</br>
 +
In addition, we also use the finite difference method on the time scale. Thus, we get:
 +
</br>
 +
</br>
 +
<center><img src="https://static.igem.org/mediawiki/2012/8/84/Eq37_grenoble.png" alt="" /></center>
 +
</br>
 +
</br>
 +
where p=∆t/Nt where ∆t is the interval of time we work in and N<SUB>t</SUB> the number of points of our time discretization.
 +
</br>
 +
</br>
 +
With these approximations, we could solve the equations.
</br>
</br>
</section>
</section>

Revision as of 15:48, 21 September 2012

iGEM Grenoble 2012

Project
              
The goal of this part was only to check the speed of the diffusion. We use the same model as the 2011 IGEM team of Grenoble, which is based on the Bangalore 2007 IGEM team model. We first have to add the diffusion terms in the equations thus we get:



Because of the temporal and special derivatives, we couldn’t use a classic matlab solver to solve this set of equations. The approximation we used consisted in dividing the space into a grid:



It is the finite difference method. We thus get:



with h=lx/N, where lx is the length of the grid and N is the number of points of discretization along x. By using the same approximation on yi, and assuming that ly=lx, and that we have the same number of points of discretization, we get:



In addition, we also use the finite difference method on the time scale. Thus, we get:



where p=∆t/Nt where ∆t is the interval of time we work in and Nt the number of points of our time discretization.

With these approximations, we could solve the equations.