This is the readme for the NEURON model from the paper:
Ferrante et al, Feed-forward inhibition as a buffer of the neuronal
input-output relation.  Proc Natl Acad Sci U S A. (2009)
In this paper, the model shows that feed-forward inhibition (FFI) can
change a steeply sigmoid I/O curve into a double-sigmoid typical of
buffer systems.  Specifically Fig_1B_left.hoc and Fig_1B_left.hoc
respectively produce plots of firing response of the model MOPP cell
(left) and granule cell (right) upon somatic current injections.
Under linux/unix systems:
to compile the mod files use the command 
and run the simulation hoc file with the command 
nrngui Fig_1B_left.hoc
Under Windows systems:
to compile the mod files use the "mknrndll" command.
A double click on the simulation file
will open the simulation window.
Under MAC OS X:
to compile the mod files drag and drop the extracted folder onto the
mknrndll icon. Drag and drop the mosinit.hoc file onto the nrngui
Once NEURON started you can set a current amplitude in the graph
textbox, press the button "init & run" and check the corresponding
output firing rate (Hz).

When the simulations for the two cells are done the results should
look similar to these:
  screenshot You can try another current amplitude and reiterate the process to build an I-f curve (see paper for details). Questions on how to use this model should be directed to