Spyke

spyke

spyke is a Python application for visualizing, navigating, and spike sorting high-density multichannel extracellular neuronal waveform data.

spyke currently works with .dat files, Blackrock .nsx files, Swindale Lab .srf and .tsf files, and Rodrigo Quian Quiroga's simulated data .mat files. spyke can be extended to work with other electrophysiology file formats. Some sample data is available. Spike sorting for polytrodes: a divide and conquer approach is a paper describing the overall approach. spyke is described in greater detail in Chapter 3 and Appendix C.2 of this thesis.

A much older version is described in the paper Python for large-scale electrophysiology.

Some functionality was inherited from Tim Blanche's Delphi program "SurfBawd". Some icons were copied from Ubuntu's Humanity icon theme.

Dependencies

spyke requires recent versions of the following to be installed:

• Python (>=3.6)
• PyQt5 (PySide will probably work too, but is untested)
• PyOpenGL
• IPython
• numpy
• scipy
• matplotlib
• Cython
• MDP
• scikit-learn
• PyWavelets
• jsonpickle
• simplejson

spyke is developed in Xubuntu 20.04. It should work in other Linux distributions, and is known to work in OSX. In principle, it should also work in Windows.

spyke is a Qt5 application. To make it look like a normal GTK application on a GTK-based desktop like (X)ubuntu, make sure to install the qt5-style-plugins package in (X)ubuntu, and then launch the qt5ct config app to set your Qt style to gtk2.

Installation

Most often, you'll want to do a "developer" install, which lets you work on or otherwise update spyke in-place, in whatever folder you cloned it into with git, while still being able to call import spyke and use it as a library system-wide. This creates an egg-link in your system site-packages or dist-packages folder to the source code:

$ sudo python setup.py develop

or the equivalent using pip:

$ sudo pip3 install -e .

This will also install a bash script on your system so that you can simply type spyke at the command line to launch it from anywhere.

$ spyke

Alternatively, you can launch spyke with:

$ python -m spyke.main

which gives you some more flexibility, such as allowing you to specify what version of Python you want to use.

Instead of development mode, you can also install by copying the code to your system Python installation:

$ sudo python setup.py install

However, unlike in developer mode, every time you update from git, you'll have to re-run the above installation command.

Documentation

See TUTORIAL.md for a fairly brief tutorial.

Many keyboard shortcuts are available. Tooltips give some hints. You can also discover them by searching for keyPressEvent methods in the code.