Running NEURON model packages
The instructions below are based on the instructions for running models from ModelDB (http://senselab.med.yale.edu/ModelDB/). They assume you have installed a recent NEURON package (>7.3 is known to work), available at http://www.neuron.yale.edu. For Mac OS X we strongly suggest to install Neuron using Homebrew (see instructions below). The current models support python 2.6+.
Linux / Unix / Mac
1. Download one of the model packages from the Downloads section, or from an me-type fact sheet. For this example we will download the package named L5_DBC_cACint209_3.zip (i.e. m-type L5_DBC, e-type cAC, exemplar 3)
2. Unzip into the appropriate location
user@machine:~/dev/sim/neuron$ unzip L5_DBC_cACint209_3.zip user@machine:~/dev/sim/neuron$ cd L5_DBC_cACint209_33. Launch the hoc GUI
user@machine:~/dev/sim/neuron/L5_DBC_cACint209_3$ sh run_hoc.sh
4. Read the 'How to use the ME-type model GUI' section below
1. Download one of the model packages from the Downloads section, or from an me-type fact sheet. For this example we will download the package named L5_DBC_cACint209_3.zip
2. Unzip the downloaded package
3. Run mknrndll and point it to the 'mechanisms' directory inside the unzipped cell package directory
4. Run nrngui and from using 'File->working dir' menu option change to the 'mechanisms' directory. Then load the 'mosinit.hoc' script from the root of the unzipped cell package directory
5. The GUI should load automatically. Read the 'How to use the ME-type model GUI' section below
How to use the ME-type model GUI
To start simulating the model, press the Init & Run button. Initially there are no stimuli present, which means that the cell will stay at its resting membrane potential.
There are three panels that show the output of the model. One is a graph with the membrane voltage recorded in the soma (units: ms and mV), the two other panels have a representation of the cell's morphology. The sections in the first morphology change color dependending on their membrane voltage during the simulation. When synaptic input is present, the second morphology will show the location of the activated synapses.
Two cell can be stimulated in two ways: step currents (current clamp) and synaptic input.
Step currents can be selected by choosing 'Step current x' using the radio buttons. Every step corresponds with the same amplitude as was used to generate the plots on the website.
Synaptic inputs can be enabled by pressing one of the buttons from the list of presynaptic m-types. When a certain m-type is selected, all the synapses that cells from the specified m-type make on the simulated cell (in the neocortical microcircuit model) will become active. Every presynaptic cell will be represented by a Poisson spike train, and the synapses will receive input from these virtual cell. The default firing rate of these virtual cells is set to 10 Hz. The user can change these value by changing the input field next to the respective m-type activation button.
Running the Python version of the model package on Linux / Unix / Mac OS X
If you have installed Neuron with Python support, you can run a python version of the model. The python script depends on numpy and matplotlib, so first install these packages with e.g. 'pip install numpy matplotlib'.
The script will run without a GUI, and will produce traces for step current 1, 2 and 3 in the directory python_recordings
user@machine:~/dev/sim/neuron/L5_DBC_cACint209_3$ sh run_py.sh
In our experience the Homebrew package manager is the most convenient way to get a commandline version of Neuron that has both X11 and Python support running on your Mac.
1. Install Homebrew by following the instruction on http://brew.sh, or:
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)")
2. Tap the Science Homebrew repo from https://github.com/Homebrew/homebrew-science, by executing:
brew tap homebrew/science
3. Now you can install Neuron:
brew install neuron
This will install all the necessary software dependencies like InterViews, X11, ...
You can test the installation by executing
which should start the Neuron GUI
To test the Neuron python module:
python -c 'import neuron'