Ephys 2.0

Electrophysiology spike-sorting pipeline, the 2nd.

• PI: Bence Olveczky, Professor of Organismic and Evolutionary Biology

• Contact: Max Shad, Anand Srinivasan, Chris Axon, Rudy Gelb-Bicknell

• Developers:

  • Mahmood Shad, Associate Director, Research Software Engineering

  • Anand Srinivasan, Sarah Leinicke and Naeem Khoshnevis, Research Software Engineers

Getting started

There are two central components to ephys2: the sorting pipeline and the GUI.

The first main choice point is to decide whether you want to run it on your local computer, or on the cluster. Unless you are working with unusually small datasets, the answer is usually the cluster.

Running on the cluster

Follow instructions in singularity/README.md

Running locally

Follow instructions in ephys2/README.md

General guidelines

These are the user docs; for source code / developer documentation, see the Python package in ephys2.

Follow these steps to use ephys2:

1. Write a configuration file; see the examples/README.md for instructions and pre-built configuration files for a variety of different processing pipelines.

2. Install ephys2 and run your configuration by either following the instructions for cluster (singularity/README.md) or local installation (ephys2/README.md).

3. Visualize and edit the results.

4. Use any of the produced datasets in your analysis, by consulting ephys2/data.md for documentation of the various HDF5 files produced by ephys2.

Project organization

• ephys2: Source Python package for Ephys2
• scripts: Utility scripts for ephys2-related workflows
• docs: Usage & API documentation
• examples: Example job configuration files
• singularity: Singularity build for running on cluster

Documentation server

http://199.94.60.144/ URL coming soon.

Resources

The FAST algorithm implemented by ephys2 is based on the 2017 eLife paper - note the Figures.