Normally the design and analysis is done/documented on the wiki, and even lab measurements/protocols are in the Notebook. This is in contrast to most of the artifacts related to modelling (SBML files, data sheets, etc.). To make our models more accessible and an integral part of our project we put the entire modelling workflow on-line. For one thing, this makes it easier to explore the model, up to the point that even non-modellers are able to explore the model.

Modelling is an integral part of synthetic biology and most of our modelling results are therefore integrated with our theoretical information and lab results on our project pages. In general we have tried to make as much of our model as possible interactively available on our wiki. Specifically, we have constructed several interactive calculators that can be used to explore our model, some including interactive graphs to show the results.

In our project we use modelling for the following purposes:

  • Description of our system. By modelling the system the different relationships between components in our system are made explicit.
  • Gaining insight in our system. Having modelled our system we can see how different variables interact, giving essential insights into how our system functions.
  • Verification of our design. For example, we looked at the number of gas vesicles needed to let our cells float, to check whether it should be possible.
  • Making design choices. We have shown that constitutive expression of ArsR can indeed significantly increase accumulation levels, and we would be able to show the impact of this constitutively expressed ArsR regulating the ars promoter on the expression of the GVP cluster (see our promoter modelling).
  • Designing tests. By looking at the behaviour of GlpF/ArsB (importer/exporter for As(III)) we determined what range of concentrations would be interesting to use in our uptake experiments.
  • Analysis of results. Using data from uptake experiments, promoter measurements and TEM pictures we can estimate further constants and/or explain the results.

Our initial ideas on how and what to model (including a survey of previously used software) can be found at Brainstorm/Modelling.


Our modelling results can be viewed on our project pages, including our interactive calculators and graphs:

Most of the modelling sections are based on our ODE model:

(Click to go to our detailed modelling page.)