GLOBUS PALLIDUS NEURON MODEL DATABASE
----------------------------------------

This file explains who to run the multi-compartmental globus pallidus
(GP) neuron model as published in:

Gunay, Edgerton, and Jaeger (2008). Channel Density Distributions Explain
Spiking Variability in the Globus Pallidus: A Combined Physiology and
Computer Simulation Database Approach. J Neurosci. 28(30):7476-7491.

Requirements:
--------------------
Genesis 2.3 from http://www.genesis-sim.org/GENESIS/.

For compressing Genesis voltage files (default), you will need the
Gencompress utility from:
http://userwww.service.emory.edu/~cgunay/pandora/#gencompress
Alternatively, you can disbale creating compressed output files by
setting the compressdata variable to 0 in setup.g.

You need to create output data directory (e.g., 'data') in the
run_sample directory for the disk_out type binary membrane voltage
data files. You can also specify other directories by modifying the
setup.g file (see below).

Running the simulation:
------------------------------

To run the model manually, the following UNIX command will run the
simulation using the maximal conductance values of the hand-tuned
model:

$ export GENESIS_PAR_ROW="500 1 .1 10 20 2 4 .3 1 1 1"
$ genesis -nox -batch -notty setup.g

See readGPparams.g for the order of parameter values.

To run it on a SGE-compatible cluster see the shell scripts below.

Data Analysis:
--------------------

See the PANDORA Toolbox for data analysis of the voltage files
(http://senselab.med.yale.edu/simtooldb/ShowTool.asp?tool=112112 or
http://software.incf.org/software/44/view/PANDORA/home/ or
http://userwww.service.emory.edu/~cgunay/pandora/).

Model Neuron Database:
------------------------------

The model database with the raw voltage data files takes about 7 GB in
compressed format and can be requested in DVD format from the
authors. The Matlab PANDORA database file is much smaller (order of
hundred MBs). Contact CG (cgunay AT emory.edu or cengique AT
users.sf.net) to request these packages.

Organization of model file hierarchy:
-----------------------------------------

DIRECTORY: common/ 

Contains common scripts to setup model channels and structure, and the morphology files.

DIRECTORY: runs/run_sample/

Contains scripts specific to running the simulations for this GP model
database.

FILES:

setup.g: Main simulation script that starts a batch of current
injection protocol simulations on a given single set of conductance
density parameters.

setup-test.g: A simpler simulation run similar to setup.g that is
intended for a test simulation.

readGPparams.g: Script to read model maximal conductance values from a
shell environment variable.

morph3_addedNaF_HCN_KCNQ_lower_vals.par: Set of 100,602 unique
parameter combinations simulated in the published GP model
database. This file contains some redundant parameter combinations,
more than actually used in the database.

paramRangesMorph*.txt: Example input files for generating the *.par
files using the paramScanSingle.pl script in this directory.


Supplemetary shell scripts under runs/run_sample/:
---------------------------------------------------------

All of the below scripts contain some documentation and examples on
how to run them within their header, or when they're run without
parameters.

sge_submit: A UNIX bash shell script shows how the original model
database was submitted to a computing cluster managed with the Sun
Grid Engine (SGE) software. 

sge_perlhash: An Perl script designed to be submitted to SGE for
running a set of simulations passing a parameters from one of the
*.par files above (See sge_submit for an example).

create_perlhash_param_db: A Perl script to prepare the *.par parameter
file to be read by sge_perlhash script (See sge_submit for an example).

run_genesis.sh: An SGE script to submit a single simulation to the
cluster without passing any parameter values.

---
Model files submitted by
Cengiz Gunay <cgunay AT emory.edu>, 2008/08/11

Other authors:
Jeremy Edgerton <jedgert AT emory.edu>
Dieter Jaeger <djaeger AT emory.edu>

See also the Jaeger Lab web page:
http://www.biology.emory.edu/research/Jaeger