Team:Technion HS Israel/Modelling/Results

• Home
  Medal Fulfillment Home
• Project More about our project
  Description Results Parts
• Documents
  Experiments Notebook safety Protocols
• Modelling We take our math seriously
  Model Overview Plasmid Loss Equations Parameters Results Documents
• Software
  Software Overview Search Engine Constants Database Simulation About the Database
• Team We are a bunch of nerdy nerds
  Members Attributions and contributions Gallery
• Practices
  Collaborations Practices Search Engine Modeling Tutorials

Modeling Results

Figure one: In this graph we see the amount of ccdb (the toxin substance)inside the bacteria as a function of time. Both the X axis and the Y axis are in arbitrary units, for a few different values of initial AHL in the cells. Each line stands for a different initial amount of AHL. The one on the top is the graph of AHL(t=0)=2 (again, in arbitrary units) and the bottom one is the graph of AHL(t=0)=6 .
We get a few insights from this set of graphs: we can see a peak at the beginning in all the graphs of ccdb, it's because of the fact that in the time t=0, there is no TetR in the cell the time it takes to the tetR to form and repress the ccdb creation . We don't want it to kill the cells. We can solve it by using modelling to determine how much AHL to put so the peak will be below the concentration that kills the bacteria. This graphs show an ideal kill switch - there is a great sensitivity to the dependent variable (AHL). That is, by changing the amount of AHL we put in the tube we get a big change in the time the bacteria will live, as long as the fetal dose of ccdb is high enough so the peak won't kill the bacteria (the peak has a low variability in its position, as you can see in the graph). It shows that our system is holding true to the flexibility promise. The system fits to a wide range of time spans.

Figure two: This is a graph of the TetR as a function of time for different values of AHL at time 0. The AHL in on the range from 2.0 to 4.0 A.U. . We did a similar experiment in the lab, as described in the Project results page. You can see that its shape is similar to the shape of the results of the experiments.

Figure three: In the following picture sequence you can see graphs of ccdb as a function of time for different values of the degradation constant of ccdb, on the range of 0.01 to 3.2 A.U.. The shape of the graph for 0.01 A.U. isn't what we want because after the peak at the beginning, the ccdb doesn't decrease (because it degrades very slowly). It tells us that our system functioning is strongly depended on the degradation constant of the toxin protein, as if it doesn't degrade fast enough a high level of ccdb will remain in the cells and kill them slowly even if the killing threshold is higher than the peak. The shape of the graphs for the lower degradation constant isn't a shape of a good autonomous kill-switch as it just increases amounts of the toxin until it kills the cell. Therefore, we should use a protein with a high degradation constant. The results of the experiments are similar to the modelling graphs of a low degradation rate. It reveals a possible artifact in our experiment: in the experiments we use ccdb but in the lab we tested our system using RFP, which has different properties, including different degradation constant, than ccdb. This is an example for the importance of modelling. It enables us to use this result to predict problems in the lab and understand better the experiments and their results.