Timothysit / spike-network-viewer

Dashboard for looking at functional connectivity from spike data.

Spike Network Viewer

This is a (work in progress) dashboard for visualising the network properties of neural activity from spikes. Note that currently the spatial structure of the network is assumed to be that of the MultiChannels 60MEA system (8 x 8 layout)

Installation Instructions

Step 1: Clone the repository

• option A: (go to a suitable direcotry, such as home), and run git clone https://github.com/Timothysit/spike-network-viewer
• option B: click on the green Code button, then do Download ZIP, then unpack the contents

Step 2: Set up python environment

1. If you use anaconda, change to the folder containing the repository contents (cd spike-network-viewer) and create a new environment using conda env create --name spike-network-viewer --file=spike-network-viewer.yml
   • feel free to replace 'spike-network-viewer' with your preferred name of the environment
2. Install the required packages by running: pip install -r requirements.txt

Step 3: Run command line argument to launch viewer

1. Change to the directory containing the cloned repoistory, eg. cd spike-network-viewer, then do:
2. Activate your python environment, eg. conda activate spike-network-viewer if you use anaconda as your python package manager
3. Launch the viewer: panel serve --show spike-network-viewer.ipynb

File format requirements

The current supported file format is limited to:

• pickle files: .pkl

An example file can be found in the example-files folder.

Converting .mat files to .pkl files

Go to the spike-network-viewer folder, and run:

python file_conversion.py --mat-folder '/folder/to/your/mat-file/'

Pickle file structure

For pickle files, the content should be a dictionary with the following structure:

• metadata: a dictionary with:
  • Name: name of the file (optional)
  • num_channels: number of channels in the recording (int)
  • num_frames: number of frames in the recording per channel (int)
  • fs: sampling rate (float)
  • channel_id: numpy array containing the names of the channels
• df: pandas dataframe with two columns:
  • spikeTime: time of spike (s)
  • electrode: electrode id (int or float)

Modifying the viewer

The viewer is contained in the file spike-network-viewer.ipynb, which can be opened using:

1. jupyter notebook, in which case run: python -m jupyter notebook
2. jupyter lab, in which case run: python -m jupyter lab

then open the notebook, which allows you modify the behaviour or add/subtract functionalities to the viewer.

Contributing

Feel free to post issues if you have trouble running the app or have feature suggestions!