Skip to content

Extended edit similarity measurement for high dimensional discrete-time series signal (e.g., multi-unit spike-train).

License

Notifications You must be signed in to change notification settings

oist-ncbc/spykesim

Repository files navigation

PyPI MIT License Build Status

spykesim is a Python module that offers functions for measuring the similarity between two segmented multi-neuronal spiking activities.

Extended edit similarity measurement is implemented. You can find details in the following paper.

https://www.frontiersin.org/articles/10.3389/fninf.2019.00039

This library is re-implementation of the algorithm. The original implementation can be found in this repo.

Supported Operating Systems

This library tested on Ubuntu and MacOS.

For Windows users: Please consider to use Ubuntu via Windows Subsystem for Linux.

Installation

If you do not have Python3.7 on your environment, you may use Anaconda.

Cython and Numpy needs to be preinstalled as these will be used in the installation process.

If you have not installed these packages, run the following:

pip install numpy cython

You can install this library via pip as well:

pip install spykesim

or you may clone and build by yourself:

git clone https://github.com/KeitaW/spykesim.git
cd spykesim
python setup.py build_ext --inplace install

Dependencies

  • Python (>= 3.7)
  • Numpy(Needs to be preinstalled)
  • Cython(Needs to be preinstalled)
  • scipy
  • tqdm
  • h5py

Tutorial

You can find a tutorial in doc.

Citation

You can use the following bib entry to cite this work:

@article{Watanabe:2019eq,
author = {Watanabe, Keita and Haga, Tatsuya and Tatsuno, Masami and Euston, David R and Fukai, Tomoki},
title = {{Unsupervised Detection of Cell-Assembly Sequences by Similarity-Based Clustering}},
journal = {Frontiers in Neuroinformatics},
year = {2019},
volume = {13},
month = may
}