Skip to content

Python package extending MNE-Python to spike analysis.

License

Notifications You must be signed in to change notification settings

labianca/pylabianca

Repository files navigation

pylabianca: Python tools for spike analysis

labianca codecov

pylabianca offers a simple and efficient way to read, analyze, and statistically compare spike data in just a few steps. Key features include:

  • A familiar API inspired by mne-python, ensuring ease of use for experienced users.
  • Two intuitive data structures - Spikes and SpikeEpochs - for organizing and storing spike data.
  • Integrated support for storing trial-level metadata, enabling easy trial selection based on conditions, similar to mne-python.
  • Outputs in the form of xarray DataArrays, which come with labeled dimensions and coordinates.
  • Seamless metadata inheritance in xarrays, allowing for visualizations by condition using pylabianca.viz.plot_shaded or native xarray plotting functions.
  • Built-in support for statistical testing via cluster-based permutation tests, facilitating comparisons between different conditions based on trial metadata.

installation

pylabianca can be installed using pip:

pip install pylabianca

To get most up-to-date version you can also install directly from github:

pip install git+https://github.com/labianca/pylabianca

what's new?

See whats_new.md for documentation of recent changes in pylabianca.

docs

Online docs are currently under construction.

Below you can find jupyter notebook examples showcasing pylabianca features.

To better understand the data formats read natively by pylabianca (and how to read other formats) see data formats page.

sample data

You can get example human data that are used in the examples here.
The preprocessed FieldTrip data used in the examples are available here.

About

Python package extending MNE-Python to spike analysis.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages