From 2009.igem.org
(Difference between revisions)
|
|
| Line 166: |
Line 166: |
| | <br> | | <br> |
| | <div class="heading"> | | <div class="heading"> |
| - | Descriptive Title of What You’re Doing
| + | Modelling Meeting |
| | </div> | | </div> |
| | <br> | | <br> |
| | <div class="desc"> | | <div class="desc"> |
| | </html> | | </html> |
| - | WIKI CODING HERE
| + | Today, we went over to ICT to discuss about how the two types of modelling can be put together as a whole. We discussed what both projects are capable of doing and the weaknesses in both models. The main task that matlab based modelling can help us achieve is to understand the signalling pathway in more depth. By characterizing the system via testing different parts of the circuit, we can understand more about the AI-2 system. If our results from the lab do not match the model, then either we are missing something in our model or there might be a step that we did not consider in the biological model. However, membrane computing gives us a better picture of the overall picture. It has the power to look at populations of cells as well as focus on one cell. I think the appropriate way to sell both modelling systems is to sell MC as a simulation tool and using the matlab based model, we can understand the signalling cascade within the cell more. There are a list of things we need to research about in the next few days for friday. We have another meeting on Friday for more discussion. |
| | <html> | | <html> |
| | </div> | | </div> |
Revision as of 22:57, 12 August 2009
University of Calgary
|
|
CAROL
Modelling Meeting
Today, we went over to ICT to discuss about how the two types of modelling can be put together as a whole. We discussed what both projects are capable of doing and the weaknesses in both models. The main task that matlab based modelling can help us achieve is to understand the signalling pathway in more depth. By characterizing the system via testing different parts of the circuit, we can understand more about the AI-2 system. If our results from the lab do not match the model, then either we are missing something in our model or there might be a step that we did not consider in the biological model. However, membrane computing gives us a better picture of the overall picture. It has the power to look at populations of cells as well as focus on one cell. I think the appropriate way to sell both modelling systems is to sell MC as a simulation tool and using the matlab based model, we can understand the signalling cascade within the cell more. There are a list of things we need to research about in the next few days for friday. We have another meeting on Friday for more discussion.
|
|
|
CHINMOYEE
Descriptive Title of What You're Doing
|
|
|
EMILY
Descriptive Title of What You're Doing
|
|
|
FAHD
Descriptive Title of What You're Doing
|
|
|
IMAN
Descriptive Title of What You're Doing
|
|
|
JAMIE
Descriptive Title of What You're Doing
|
|
|
JEREMY
Descriptive Title of What You're Doing
|
|
|
KATIE
Descriptive Title of What You're Doing
|
|
|
KEVIN
Descriptive of What You're Doing
|
|
|
MANDY
Descriptive Title of What You're Doing
|
|
|
PATRICK
Descriptive Title of What You're Doing
|
|
|
PRIMA
Descriptive Title of What You're Doing
|
|
|
STEFAN
Descriptive Title of What You're Doing
|
|
|
VICKI
Meeting with the modelling team
The modelling crew met today to discuss how we could integrate our work into a cohesive section in a presentation. Our main focal points were (a) why model; (b) what questions can be answered by each modelling approach; (c) what does each modelling approach NOT answer; and (d) can we resolve the gaps in one modelling approach by bringing in the other? This will help frame our descriptions of our approaches and help us convince others that what we're doing is worth it.
On the MatLab side, we have found that SimBiology does not provide the simulation power that can be achieved by membrane computing. For instance, the platform does not offer a convenient way of modelling more than one cell, making it very difficult to introduce a space/location component into our model. We are having trouble resolving what questions can be anwered with the SimBiology simulation model that cannot be answered by using membrane computing. Accordingly, we are focussing more heavily on characterisation. As we will be biobricking ~5 different promoter-LuxPQ components, each one of those circuits can be contributed to the registry in characterised form, so that future users will be able to use our characterisation data to compare parts. It is sensible to approach this with MatLab because the membrane computing model cannot answer many of these aspects - for example, what is the best temperature to culture and maintain our cells for optimal performance? Or how does the broth in which our cells are cultured affect the dynamic performance? By collecting that data and fitting it to equations, future users of our registry contributions will have a quantitative way to compare parts and select the best one for their purposes.
Our next meeting will be at noon on Friday. By then, this is what I hope to achieve:
- play with SimBiology at the GUI and command line levels to see if we can build in a multicellular approach
- check out the optimisation toolbox and see if there would be a nice way to implement some form of evolutionary optimisation in MatLab. This will be useful in helping us optimise reaction constants for the simulation model. Work with Afshin on this - if it can't be done on Matlab, we might be able to implement something in mathematica using the same data.
- work with the Fort Mac crew to prepare a cohesive, relevant presentation on our experience on the oilsands tour and how synthetic biology can tie into that. Note that iGEM also focusses on ethics and social responsibility, which will be a critical consideration in forging any future alliances, regardless of the industry with which our potential partners are affiliated.
|
"
