Introduction to using the Traub et al 2005 model¶
First, and most importantly, please remember that this is a work in progress! If you would like to help make this model more useful for the community, please get in contact via the OSB Discuss mailing list.
The original Traub et al model was developed in FORTRAN, this was converted to NEURON by Tom Morse and Michael Hines , and this has now been converted to NeuroML & neuroConstruct.
Please make sure to read about the Known issues with this model.
Important details of the process of conversion of the cell models to NeuroML, and matching cell behaviour across simulators is present in the 2010 NeuroML paper.
Install neuroConstruct & get latest project¶
See the instructions here regarding obtaining the latest version of neuroConstruct.
Install NEURON, GENESIS and/or MOOSE (see project:simulators).
To get a local copy of the Thalamocortical project, install Git and type:
git clone https://github.com/OpenSourceBrain/Thalamocortical.git
View a cell in 3D¶
Open neuroConstruct and open the Thalamocortical project; File -> Open Project -> select
Go to tab Visualisation, select L23PyrFRB, a Fast Regular Bursting Layer 2/3 Pyramidal cell, in the drop down box (or any other cell of your choosing) and click View.
To change the view so that all segments are drawn as solid cylinders select All solid in the lower left drop down box.
To see the distribution of channels on the cell, select Cell density mechanisms in the lower right drop down box.
To get more details on the properties of this cell, go to tab Cell Types, select L23PyrFRB, and a summary of the biophysical properties and structure of the cell is given.
Test installation with single cell simulation¶
Assuming the simulator NEURON is installed, a simulation can be generated with just this cell. Go to tab Generate and select Simulation Configuration: Cell2-suppyrFRB-FigA1FRB in the drop down box. This will generate a network with one L23PyrFRB and current clamp input (hyperpolarising followed by depolarising).
Go to tab Export, sub tab NEURON, and click Create hoc simulation. This generates the HOC and MOD files specific to NEURON for this cell & network steup. These files can be viewed using View…. Execute the simulation in NEURON using Run Simulation.
This simulation can be loaded back into neuroConstruct. Go to tab Visualisation, click View prev sims in 3D, select the simulation and press Load simulation.
To plot the data saved in the simulation, click on any part of the cell, and click Plot selected. This will list the locations recorded (soma and one point on axon).
To replay the simulation in the 3D view select Replay. This will give a time varying coloring to the cell based on membrane potential. Note that since only 2 locations are recorded, the majority of the cell will be coloured according to the membrane potential at the soma.
Generate synapse files (important!)¶
Currently the majority of the synapse models (ChannelML based synapses in cellMechanism folder) are not included in the project in the repository.
These are generated by a script makeSyns.sh in pythonScripts/netbuild. This script was manually created from the info in teh Appendix of the original paper (it’s easier to check/update the values in makeSyns.sh than through the neuroConstruct GUI).
This script (or makeSyns.bat on Windows) needs to be run once to generate files cellMechanisms/Syn_AMPA_L4SS_L4SS/ChannelMLSyn.xml etc.
Generate a network model¶
Once the synapse mechanisms have been generated, it is possible to generate a 3D network model containing all 14 cell types from the original paper.
As this network example is quite large, it is wide to view the cells with only lines for neurites. Go to menu option Settings -> Project Properties and Soma solid, neurite lines for the Display option. Press Save to exit.
Go to tab Generate and select Simulation Configuration: Demo3D in the drop down box. This will generate a network with 14 cell groups. Go to tab Visualisation, select Latest Generated Positions and click View.
Click on one of the cells (or select a cell group/cell number in the drpdown boxes) and click Transparent mode to better see the position of individual cells (and the somata of cells it’s connected to):
Please make sure to read about the Known issues with this model.