Paris/Modeling general



French Page

What do we want to do ? The goal is to model the cellular growth, and to prove it is possible, to find the range of paramaters that enables it, and then to optimise the triglycerid production. Is it Possible for a cell producing the Amino Acide, to feed at the same time the mother cell, and the triglyciride factory one? For this, we need to find the good theoritical balance between the different elements of the population. This balance is determined by the firing rate of the cre/lox system, which is genetically engineered


Why a modelization ?

There is a nice discussion on the Imperial'2006 project page on how the experimental project and the modelization worked together. It is a predator-prey project and they seem to claim that it worked for them...

Minimise the biological steps

Make the system work by finding the possible parameters, and save time when doing wetwork, in order to directly make make the construct the most likely to work. ( For instance, immediately use the good promoters or RBS) Some might say the system is "simple enough" to build, so that it can be hand crafted. It might work so. But if we don't try to model a simple system, and try to make it stick as much as possible to reality, how can we ever hope to do it for a more complex one?

Biosynthetic spirit IGEM

If you listen toEndy,biosynthetic is not making a biological construct. It's HOW you make it, it's the process you use to build it. It's giving the possibility to first virtually create the system before carrying it out: by assembling the biobricks as judiciously as possible, by exactly knowing the link between input and ouput. It has to be reflected in the construction process.

Rythme and organise the biological constructs

We need to characterise the biobricks we're about to use, BEFORE assembling them, so that we can feed the model with parameter values. The We use the model to construct, and see if it fits or not the model, and try to understand why. It's a new way of research in biology : instead of tearing complicated system apart, you try to first build simple ones, and by observing the difference with theory, a new explanation will sharpen the model.

Experimental constraints

Before we start, we will need to understand what are the experimental constraints, our resources and what are the essential issues for the success of our project: modelling is important by itself, but the optimal would be that modelling and wetwork pursuit the same goals and help each other. This will allow us to be performant as soon as possible.

These are some important things to know (among others) before we start:

  • What are the most important QUESTIONS that modelization can answer?
  • What are the CHOICES to be made along the project?
    • Choice of promoters?
    • Choice of circuits?
    • Choice of experimental conditions? Initial conditions? Concentrations?
  • What can we measure?
    • How long does it take to measure it?
      • What resources does the measurement need?
  • At what moment in the project planning will we be able to measure it?
  • What do we do if we cannot know (on time/never) the value of an important variable?
    • Literature search
    • Parsimony
  • Scales:
    • Lifetime of a normal cell
    • Lifetime of a cell that is missing a key amino acid.
    • Size of a normal cell
    • Size of a spaghetti cell

Model WHAT ?

Macroscopic Modelization

By macroscopique, I mean the modelization at cellular cultur level, with parameter that you can measure outside the cell. We need to get the parameter input and output values of the different elements of the system, introduce them in the model, get the desired cre lox transformation rate, and modifie the cre promoteur, or it's RBS by using caracterised biobricks.

Parameters to look for in the litterature then to measure in vitro by ourselves :

  • Absorbtion of the vital AA by a cell
  • Production of the vital AA by a cell
  • Diffusion zone of the AA
  • Lethal width of a cell without FTSZ ( for a AA producing cell) (how do you measure it? cell cycle? 4 OR 5 says Ariel. Same cycle speed as for a normal cell (without FTSZ)?
  • Lethal width of a cell without FTSZ ( for a triglyceride producing cell)?
  • Cre lox recombination rate

Always try to use a reference promoteur or RBS for the caracterisation

Microscopique or Genetic Modelization

It's the modelisation of what happens IN the cell : of the genetic network, with the quantity of the different proteins as parameter How can we measure the moduls'efficiency? Thanks to Pops and Rips (Ribosome per seconde). We need reference unity measures. For the Ribosome binding site, we should always compare the RBS used to this one. Is it also possible to have a reference promotor?