CoAxLab / CBGTControlEnsembles

Identifying control ensembles for information processing within the cortico-basal ganglia-thalamic circuit

CBGT

This repository contains code for implementing the spiking cortico-basal ganglia-thalamus (CBGT) network and drift-diffusion model (DDM) fits described in the manuscript Identifying control ensembles for information processing within the cortico-basal ganglia-thalamic circuit.

The code requires several dependencies to be installed (see below for instructions). After completing the installation procedure below, the demo notebook can be downloaded and opened inside Jupyter.

Requirements

• OSX or Linux

• Anaconda with Python 3.* (for OSX, Linux)

• gcc (if Linux) or gcc-8 (if OSX, see here)

Installation Instructions

# create a new conda environment with python 3.6
# and hit 'y' to verify the install 
conda create -n cbgt_env python=3.6 anaconda ipykernel

# activate 'cbgt_env' environment
source activate cbgt_env

# use conda (not pip) to install pymc
conda install pymc=2.3.6 --no-deps

# install hddm and kabuki
pip install --upgrade kabuki hddm

# finally install numpy version 1.11.3
# (avoids hddm incompatibility with later numpy)
pip install numpy==1.11.3

# install cbgt package
pip install -U cbgt --no-cache-dir

• After installing everything, run jupyter notebook in your terminal to start Jupyter in your browser
• Drag/drop the demo notebook (CBGT_PLOSCompBio2019_Demo.ipynb) into the Jupyter browser window

Miscellaneous:

• HDDM resources: (methods paper, demo, how-to)
• Jupyter Notebook tutorial