I would like to use the multiple run fitter to optimize a model with constraints from multiple experiments at once. The tutorial "Multiple Run Fitter tutorial" suggests this is possible, but does not go through an example (its very helpful, none-the-less :-) ).
I can set up multiple generators with different waveforms in them, but I'm not sure how to run a separate protocol for each one in the same "run" (i.e. use a linear circuit to clamp my cell with three different current steps, and three different corresponding series resistances, each optimized by a corresponding experimentally recorded waveform).
Thanks!
Using the Multiple Run Fitter to optimize multiple protocols
A short answer (no example)
Every generator gives you the possibility to define a protocol constant.
Steps:
1. display the generator
2. go to the the fitness button
3. select protocol constant (now you get a picker window)
4. select the variable you want to change, e.g. the amp member variable of an instance of IClamp (myIClamp.amp). ( now you automatically return to the generator window but it doesn't show the change)
5. close the generator window
6. bring the generator window back up from the MulRunFitter window (now you should see the protocol constants).
Now you can use the clone function in the MUlRunFitter window to obtain several generators which use the same protocol constants, which you can then change to fit the the experimental values. Of course you should also add the correct experimental data.
Remark: You should make sure your protocol constant are publically accesible, watch the command line for warnings!
Steps:
1. display the generator
2. go to the the fitness button
3. select protocol constant (now you get a picker window)
4. select the variable you want to change, e.g. the amp member variable of an instance of IClamp (myIClamp.amp). ( now you automatically return to the generator window but it doesn't show the change)
5. close the generator window
6. bring the generator window back up from the MulRunFitter window (now you should see the protocol constants).
Now you can use the clone function in the MUlRunFitter window to obtain several generators which use the same protocol constants, which you can then change to fit the the experimental values. Of course you should also add the correct experimental data.
Remark: You should make sure your protocol constant are publically accesible, watch the command line for warnings!