SpikeInterface
Spikeinterface simplifies the process of running multiple spike sorters on the same dataset
Description
One pipeline, many sorters
The algorithms available for spike sorting have proliferated in recent years. While it’s great to have many options, we still don’t know how these sorters stack up against each other. It is clear that they don’t always give the same results so having an easy way to run them side-by-side is essential.
SpikeInterface, created through a collaboration between the University of Oslo, the University of Edinburgh, the Flatiron Institute, the Lyon Neuroscience Research Center, and the Allen Institute, simplifies the process of running multiple sorters on the same dataset. With a few lines of code and regardless of the underlying data format, users can: run, compare, and benchmark most modern spike sorting algorithms; pre-process, post-process, and visualize extracellular datasets; validate, curate, and export sorting outputs; and more.
SpikeInterface is open source and implemented in Python 3. The package is available on PyPI (https://pypi.org/project/spikeinterface).
Features
Features
Easy-to-use API
Modular design
Efficient use of memory mapping and parallelization
Compatible with many input and output formats
Pre- and post-processing functions for extracellular datasets
Quality metrics for validation and automatic curation
Evaluation for both ground-truth and non-ground-truth datasets
Fully reproducible analysis workflows
Compatibility with Phy
Compatible sorters
Klusta
Mountainsort4
SpyKING Circus
Kilosort
Kilosort2
HerdingSpikes2
Tridesclous
IronClust
Wave Clus
