eegkit / rgc-classification

Non-parametric physiological classification of retinal ganglion cells

Non-parametric physiological classification of retinal ganglion cells

by Jonathan Jouty (1), Gerrit Hilgen (2), Evelyne Sernagor (2), Matthias H. Hennig (1)

(1) Institute for Adaptive and Neural Computation, University of Edinburgh, School of Informatics, 10 Crichton St, Edinburgh, EH8 9AB, United Kingdom (2) Institute of Neuroscience, Newcastle University, Newcastle, NE2 4HH, United Kingdom

This notebook reproduces the analysis of the RGC data set, as shown in this preprint.

If you have comments, suggestions or would like to suggest improvements, please submit a Github issue or make a pull request.

Instructions

This code was developed using Python 3.6. Required standard python dependencies are: numpy, matplotlib, scipy, sklearn, h5py, seaborn and joblib. Missing packages should be installed using pip install package_name.

Also required is pyspike by Mario Mulansky and Thomas Kreuz. This can be installed using pip install pyspike.

To run the demo notebook, download the relevant data files (including some pre-computed data) from this address, with password rgc: https://datasync.ed.ac.uk/index.php/s/2xU82QRb2rPxv8k