Spelt - the Spatial Electrophysiology Toolbox

Spelt is a python-based toolkit for analysis of co-recorded electrophysiological and behavioural data, based on scaNpix, and using SpikeInterface v0.101+

Currently this is implemented for loading, processing and analysing raw data from Axona systems, with Neuropixels 2 Open Ephys recordings as a WIP

• The package code requires Python 3.9, owing to several dependencies

• Development by Jake Swann @ Wills-Cacucci Lab, UCL

Usage

This repo is currently designed to be used with the following workflow:

To extract and analyse CA1 pyramidal cell spike data:

1. Record raw Axona or OpenEphys files
2. Run Spike Sorting From Sheet.ipynb (requires correctly formatted Google Sheet and data directory, or you can load your own .xlsx). This collates individual trials into sessions, runs Kilosort 2/4 (depending on probe type) and extracts position data
3. Manually curate spikes in phy
4. Use the spelt.ephys object and methods to load your data into the ephys object
5. Use the analysis and mapping functions to solve the brain