Spike Sorting Pipeline

• This project was part of the Reylab spikesorting pipeline.

The following is a processing pipeline for analyzing neural recordings using MATLAB and Python. Here are the requirements for each:

MATLAB Requirements

• MATLAB version 9.6 (R2019a) or later
• Signal Processing Toolbox
• DSP System Toolbox
• Parallel Computing Toolbox
• MATLAB Parallel Server
• Polyspace Bug Finder
• NPMK for Blackrock recordings
• Neuroshare for Ripple recordings

Python Requirements

You can install the required Python packages using the following command:

pip install -r requirements.txt

Steps

1. Parsing the Recording

   • For Ripple recordings, use the parse_ripple.m script.
   • For Blackrock recordings, use the parse_NSx.m script.
   • Note: The Neuroshare library is required for Ripple recordings.

2. Analyzing Power Spectrum and Calculating Notches

   • Use the new_check_lfp_power_NSX.m script to generate figures for analyzing the power spectrum and calculating notches.
   • If you prefer to use Python for this step, there is a function implemented in Python called filter_freq_peaks that performs notch filtering. If you used the new_check_lfp_power_NSX.m script, you can set load_mat_notches=False in the Python pipeline.

3. Configuration Setup

   • Before starting the pipeline, make sure to set up the config.Yaml file in the config folder.
   • Update the paths in the configuration file to match the paths on your computer.

4. Execution

   • Open the main.ipynb notebook and follow the cells to execute the processing pipeline.
   • The notebook contains comments specifically for new Python users and provides details on how to use the spikeinterface library.

Notes We need to change the bandpass filter from butterworth to