Introduction

This is the code for biological neural network model published in the paper Beta-Band Resonance and Intrinsic Oscillations in a Biophysically Detailed Model of the Subthalamic Nucleus-Globus Pallidus Network.

The main branch is the original code written for python 2.7. The updated code for Python 3.X is maintained on the branch python3, with help of @DavidCromp.

Installation

These installation instructions are for the Conda Python ditribution.

# BluePyOpt for various simulation tools
git clone https://github.com/BlueBrain/BluePyOpt.git
cd BluePyOpt && git checkout 1456941abe425b4a20fb084eab1cb6415ccfe2b8
pip install -e .
cd ..

# PyNN for network simulation (patched version)
git clone -b lkmn-multicomp https://github.com/lkoelman/PyNN.git
cd PyNN/pyNN/neuron/nmodl && nrnivmodl
cd ../../.. & pip install -e .
cd ..

# Neo electrophysiology data formats (patched version with MATLAB annotation support)
pip uninstall neo
git clone -b lkmn-dev https://github.com/lkoelman/python-neo.git
cd python-neo 
pip install -e .
cd ..

# Tools for LFP simulation
git clone https://github.com/lkoelman/LFPsim.git
cd LFPsim/lfpsim && nrnivmodl && cd ..
pip install -e ./LFPsim


# This network model
git clone https://github.com/lkoelman/stn-gpe-model-frontiers
cd stn-gpe-model-frontiers
python setup.py develop # or: pip install -e path_to_repo
python setup.py mechanisms # Build NMODL files automatically

NOTE: the custom PyNN classes in this repository have only been tested with:

• PyNN 0.9.2 (commit id 4b80343ba17afd688a92b3a0755ea73179c9daab)
• BluePyOpt version 1.5.35 (commit id 1456941abe425b4a20fb084eab1cb6415ccfe2b8).

Running the model

The network model can be found in the subdirectory bgcellmodels/models/network/LuNetStnGpe

The commands for running the model can be found in the docstring on top, in the file bgcellmodels/models/network/LuNetStnGpe/model_parameterized.py. All possible command-line arguments can be found at the bottom of this file.

To run using MPI, use the mpirun or mpiexec command:

mpirun -n 6 python model_parameterized.py --scale 0.5 --dur 3000 --dnorm --nolfp --seed 888 --transient-period 0.0 --write-interval 3000 -id CALDNORM --outdir ~/storage/LuNetStnGpe --config configs/syn-V18_template.json 2>&1 | tee CALDNORM.log

To run from the IPython shell, use the %run magic function. This lets you inspect the model interactively after running and is similar to ipython -i <script> -- <args>:

%run model_parameterized.py --scale 0.5 --dd --dur 100 --seed 888 --transient-period 0.0 --write-interval 1000 --no-gui -id test1 --outdir ~/storage --config configs/syn-V18_template.json

Scripts to run the model on SLURM or PBS-based computing clusters are also provided:

• LuNetStnGpe/batchjobs/slurm_generate_jobs.sh will submit SLURM jobs with various parameter combinations
  • it submits the job script LuNetStnGpe/batchjobs/slurm_runjob_netsim.sh
• LuNetStnGpe/batchjobs/pbs_generate_jobs.sh will submit PBS (OpenPBS/Torque) jobs with various parameter combinations
  • it submits the job script LuNetStnGpe/batchjobs/pbs_runjob_bgnetmodel.sh

BluePyOpt optimizations

# For Parallel BluePyOpt:
# In another terminal session, create ipyparallel instances in new environment
source activate neuro
cd ~/workspace/bgcellmodels
ipcluster start

Install Supporting Tools (optional)

NEURON Syntax Definitions

NEURON Syntax definitions for Hoc and NMODL languages for Sublime Text and for VS Code.

nb_conda

Allows you to select a conda environment as Jupyter kernel.

conda install nb_conda

To enable an environment as a kernel, install the module ipykernel in it:

conda install -n my_environment ipykernel

nbstripout

This facilitates working with git and Jupyter notebooks by stripping output cells before commits. This avoids committing large binary outputs like images embedded into notebooks.

pip install --upgrade nbstripout
cd bgcellmodels
nbstripout --install

Jupyter extensions

pip install jupyter_contrib_nbextensions
jupyter contrib nbextension install --user

Now extensions can be enabled via the tab 'Nbextensions' in Jupyter. For example, 'Table of Contents' will add a handy navigation widget when working with notebooks.

---