This repository contains the datasets described in the publication Brochier et al. (2018). Massively parallel recordings in macaque motor cortex during an instructed delayed reach-to-grasp task, Scientific Data, 5, 180055. http://doi.org/10.1038/sdata.2018.55

sprenger a6d508be09 Merge branch 'to_nix' of INT/multielectrode_grasp into master 6 months ago
code abb20f0df6 Add pipenv environment and lock file 6 months ago
datasets_blackrock 1541a7e8fa Removed value origin for non-existing values (harmonized odMLs) 8 months ago
datasets_matlab d1f7f85689 Initial commit of data and code of the Reach-to-Grasp Experiment. 6 years ago
datasets_nix 5389a5f01a gin commit from PF3RD7D6 8 months ago
.gitignore ec6621acba exclude more example plots from commits 1 year ago
LICENSE.txt 875bfa398d Extended central LICENSE.txt 6 years ago
README.md c925bd89b1 Adjusted examples for blackrock and nix 9 months ago
datacite.yml d8623d7d81 Add references to scidata publication 5 years ago

README.md

Massively parallel multi-electrode recordings of macaque motor cortex during an instructed delayed reach-to-grasp task

Summary

We provide two electrophysiological datasets recorded via a 10-by-10 multi-electrode array chronically implanted in the motor cortex of two macaque monkeys during an instructed delayed reach-to-grasp task (Brochier et al., 2018). The datasets contain the continuous measure of extracellular potentials at each electrode sampled at 30 kHz, the local field potentials sampled at 1 kHz and the timing of the online and offline extracted spike times. It also includes the timing of several task-related and behavioral events recorded along with the electrophysiological data. Finally, the datasets provide a complete set of metadata structured in a standardized format. These metadata allow easy access to detailed information about the datasets such as the settings of the recording hardware, the array specifications, the location of the implant in the motor cortex, information about the monkeys, or the offline spike sorting. The two datasets can be exploited to address crucial issues in neurophysiology such as: What are the principles of neural interactions in a local cortical network and how are these interactions modulated during a well-described behavioral task? How different neuronal signals such as single-unit activity, multi-unit activity or LFPs relate to each other? Which spike sorting methods provide the best estimate of single unit activity?

Downloading the data

Using gin

Create an account on gin and download the gin client as described here. On your computer, log in using

gin login

Clone the repository using:

gin get INT/multielectrode_grasp

Large data files will not be downloaded automatically. To get them, use

gin get-content <filename>

Downloaded large files will be locked (read-only). You must unlock the files using

gin unlock <filename>

To remove the contents of a large file again, use

gin lock <filename>
gin remove-content <filename>

See here for detailed information on how to use gin.

Using git annex

Make sure git and git-annex are installed on your computer. Create an account on gin and upload your public SSH key to your gin profile. Then clone the repository using

git clone git@gin.g-node.org:/INT/multielectrode_grasp.git

Large data files will not be downloaded automatically. To get them, use

git annex get <filename>

Downloaded large files will be locked (read-only). You must unlock the files using

git annex unlock <filename>

To remove the contents of a large file again, use

git annex --force lock <filename>
git annex drop <filename>

See the git annex documentation for details.

Using the web browser

Download the latest release as a zip file by clicking on Releases on the main page at https://web.gin.g-node.org/INM6/multielectrode_grasp. This zip file will contain all small (text) files only, while large data files will not be downloaded automatically and an empty placeholder will be put in their place. To get the full content of such a large file , download these files individually as needed from the web interface by clicking on them in the repository browser.

Repository structure

Directory datasets_blackrock

Contains the two original data sets i140703-001 and l101210-001. Original data files are provided in the Blackrock file format (.nev, .ns2, .ns5, .ns6, .ccf), e.g., i140703-001.nev, i140703-001.ns6,.... The files i140703-001-03.nev and l101210-001-02.nev contain offline spike sorted data for both datasets as opposed to the original recordings i140703-001.nev and l101210-001.nev which contain the same spikes, but unreliable sorting that should not be used. The files i140703-001.odml and l101210-001.odml contain extensive metadata describing the datasets in the odML format. The Excel files i140703-001.xls and l101210-001.xls contain the same information as in the odML for easy reading and browsing, however, they are not used by the loading routines. The file odml.xsl is an XML schema that is required for viewing the odML files with a web browser. the file example_blackrock.py in the code subdirectory contains an example on using these nix files.These datasets can be

Directory datasets_nix

Contains a ready to use version data sets i140703-001 and l101210-001 in the Nix data format using a Neo structure. For practical purposes, we suggest using these files (instead of the original Blackrock files in 'datasets_blackrock) for easier access to the data. Datasets can be loaded via the Neo command

import neo
with neo.NixIO(nix_filename, mode='ro') as io:
    block = io.read_block()

The resulting Neo objects are fully annotated and processed, similar to what the custom loading code in the code subdirectory will do.

In addition, l101210-001.nix will contain a downsampled version of the raw 30Khz data (i.e., the LFP) that is comparable to the online (hardware) downsampled ns2 LFP data that is supplied with i140703-001. Here, a 4th order, 250 Hz low pass Butterworth filter was applied using phase preservation and SOS, following a 30-fold downsampling. The structure of both files therefore becomes nearly identical.

Next to l101210-001.nix and l101210-001.nix, the directory also contains the files l101210-001_no_raw.nix and l101210-001_no_raw.nix. These files do not contain the raw electrode signals sampled at 30kHz, and are therefore considerably more light-weight in terms of file size.

The code to produce the Nix files from the source files in the dataset directory is given in convert_to_nix.py in the code subdirectory. Also, the file example_nix.py in the code subdirectory contains an example on using these nix files.

Directory datasets_matlab

Contains the data and metadata output of the Python loading routines in the MATLAB .mat file format. These files are provided for convenience for MATLAB users, however, note that these files are not the original data files and contain a condensed, interpreted subset of the original data. Due to size restrictions of the MATLAB file format, the files i140703-001_lfp-spikes.mat and l101210-001 _lfp-spikes.mat contain only spikes and LFP data (for monkey N), while raw data is saved separately for each channel in correspondingly named files.

Directory code

Contains example code to help in loading and analyzing the data based on the original data files contained in the datasets_blackrock folder. The file example_blackrock.py is a Python script that acts as a tutorial for loading and plotting data. Moreover, the file example_nix.py contains the same example using the nix files stored in the folder datasets_nix. The scripts data_overview_1.py and data_overview_2.py reproduce the plots of the data found in the publication. The files neo_utils.py and odml_utils.py contain useful utility routines to work with data and metadata. Finally, the file example_matlab.m contains a rudimentary MATLAB script demonstrating how to use the data provided in the .mat files.

To run the Python example code, download the release of this repository, and install the requirements in code/requirements.txt. Then, run the example via

   cd code
   python example_blackrock.py

or (the preferred option that does not require custom code in code/reachgraspio):

   cd code
   python example_nix.py

The script produces a figure saved in three different graphics file formats.

Also, the file convert_to_nix.py contains code that produces the easy-to-use Nix files (in datasets_nix) from the original source data (in datasets_blackrock).

Directory code/reachgraspio

Contains the file reachgraspio.py, which contains the loading routine specific to the Reach-to-Grasp experiments in this repository. This loading routine merges the recorded data with metadata information from the odML files into a common Neo object. It is recommended that this loading routine is used in combination with the odML and Neo libraries (see below) to work on the data.

Python dependencies

The provided code is based on the Python libraries Neo[1], odML[2], and Elephant[3], that are required by the example scripts and the reachgraspio loading routine. In short, Neo provides the data model, generic Blackrock loading routines, and APIs used to load the data; odML provides an API to handle the metadata files; and Elephant is a library for the analysis of neuronal data based on the Neo data model that is used by the example script for filtering raw signals to obtain offline filtered LFPs. To run the provided code install the Python dependencies listed in code/requirements.txt. Note however, that the versions specified in the code/requirements.txt are fixed to match the provided version of thereachgraspio.py loading routine (see Updates below) and more recent versions might exist.

Updates

Updated versions of the codes will be provided at: https://web.gin.g-node.org/INT/multielectrode_grasp This includes, in particular, the loading routine reachgraspio.py, which may need to be adapted as new versions of the Neo and odML libraries become available.

Related Publications

  • Brochier, T., Zehl, L., Hao, Y., Duret, M., Sprenger, J., Denker, M., Grün, S. & Riehle, A. (2018). Massively parallel recordings in macaque motor cortex during an instructed delayed reach-to-grasp task, Scientific Data, 5, 180055. http://doi.org/10.1038/sdata.2018.55
  • Riehle, A., Wirtssohn, S., Grün, S., & Brochier, T. (2013). Mapping the spatio-temporal structure of motor cortical LFP and spiking activities during reach-to-grasp movements. Frontiers in Neural Circuits, 7, 48. https://doi.org/10.3389/fncir.2013.00048
  • Milekovic, T., Truccolo, W., Grün, S., Riehle, A., & Brochier, T. (2015). Local field potentials in primate motor cortex encode grasp kinetic parameters. NeuroImage, 114, 338–355. https://doi.org/10.1016/j.neuroimage.2015.04.008
  • Torre, E., Quaglio, P., Denker, M., Brochier, T., Riehle, A., & Grun, S. (2016). Synchronous spike patterns in macaque motor cortex during an instructed-delay reach-to-grasp task. Journal of Neuroscience, 36(32), 8329–8340. https://doi.org/10.1523/JNEUROSCI.4375-15.2016
  • Zehl, L., Jaillet, F., Stoewer, A., Grewe, J., Sobolev, A., Wachtler, T., Brochier, T., Riehle, A., Denker, M., & Grün, S. (2016). Handling Metadata in a Neurophysiology Laboratory. Frontiers in Neuroinformatics, 10, 26. https://doi.org/10.3389/fninf.2016.00026
  • Denker, M., Zehl, L., Kilavik, B. E., Diesmann, M., Brochier, T., Riehle, A., & Grün, S. (2017). LFP beta amplitude is predictive of mesoscopic spatio-temporal phase patterns, 1703.09488 [q-NC]. https://arxiv.org/abs/1703.09488

Licensing

Creative Commons License
Massively parallel multi-electrode recordings of macaque motor cortex during an instructed delayed reach-to-grasp task in the directories datasets and datasets_matlab by Institut de Neurosciences de la Timone (INT), UMR 7289, CNRS – Aix Marseille Université, Marseille, France and Institute of Neuroscience and Medicine (INM-6), Forschungszentrum Jülich, Jülich, Germany is licensed under a Creative Commons Attribution 4.0 International License.

All code in the directories code and code/reachgraspio are each published under the BSD 3 clause licenses. See the LICENSE.txt or LICENSE files in the corresponding directories for the full license.

datacite.yml
Title Massively parallel multi-electrode recordings of macaque motor cortex during an instructed delayed reach-to-grasp task
Authors Brochier,Thomas;Institut de Neurosciences de la Timone (INT), UMR 7289, CNRS – Aix Marseille Université, Marseille, France;orcid.org/0000-0001-6948-1234
Zehl,Lyuba;Institute of Neuroscience and Medicine (INM-6) and Institute for Advanced Simulation (IAS-6) and JARA BRAIN Institute I, Jülich Research Centre, Jülich, Germany;orcid.org/0000-0002-5947-9939
Hao,Yaoyao;Institut de Neurosciences de la Timone (INT), UMR 7289, CNRS – Aix Marseille Université, Marseille, France;orcid.org/0000-0002-9390-4660
Duret,Margaux;Institut de Neurosciences de la Timone (INT), UMR 7289, CNRS – Aix Marseille Université, Marseille, France;orcid.org/0000-0002-6557-748X
Sprenger,Julia;Institute of Neuroscience and Medicine (INM-6) and Institute for Advanced Simulation (IAS-6) and JARA BRAIN Institute I, Jülich Research Centre, Jülich, Germany;orcid.org/0000-0002-9986-7477
Denker,Michael;Institute of Neuroscience and Medicine (INM-6) and Institute for Advanced Simulation (IAS-6) and JARA BRAIN Institute I, Jülich Research Centre, Jülich, Germany;orcid.org/0000-0003-1255-7300
Grün,Sonja;Institute of Neuroscience and Medicine (INM-6) and Institute for Advanced Simulation (IAS-6) and JARA BRAIN Institute I, Jülich Research Centre, Jülich, Germany;orcid.org/0000-0003-2829-2220
Riehle,Alexa;Institut de Neurosciences de la Timone (INT), UMR 7289, CNRS – Aix Marseille Université, Marseille, France
Description We provide two electrophysiological datasets recorded via a 10-by-10 multi-electrode array chronically implanted in the motor cortex of two macaque monkeys during an instructed delayed reach-to-grasp task. The datasets contain the continuous measure of extracellular potentials at each electrode sampled at 30 kHz, the local field potentials sampled at 1 kHz and the timing of the online and offline extracted spike times. It also includes the timing of several task and behavioral events recorded along the electrophysiological data. Finally, the datasets provide a complete set of metadata structured in a standardized format. These metadata allow easy access to detailed information about the datasets such as the settings of the recording hardware, the array specifications, the location of the implant in the motor cortex, information about the monkeys, or the offline spike sorting.
License CC-BY (http://creativecommons.org/licenses/by/4.0/)
References Brochier, T., Zehl, L., Hao, Y., Duret, M., Sprenger, J., Denker, M., Grün, S. & Riehle, A. (2018). Massively parallel recordings in macaque motor cortex during an instructed delayed reach-to-grasp task, Scientific Data, 5, 180055. [] (IsPartOf)
Zehl, L., Jaillet, F., Stoewer, A., Grewe, J., Sobolev, A., Wachtler, T., … Grün, S. (2016). Handling Metadata in a Neurophysiology Laboratory. Frontiers in Neuroinformatics, 10, 26. [] (HasMetadata)
Riehle, A., Wirtssohn, S., Grün, S., & Brochier, T. (2013). Mapping the spatio-temporal structure of motor cortical LFP and spiking activities during reach-to-grasp movements. Frontiers in Neural Circuits, 7, 48 [] (HasMetadata)
Funding Helmholtz Association, Supercomputing and Modeling for the Human Brain
EU, EU.604102
EU, EU.720270
DFG, DFG.GR 1753/4-2
DFG, DFG.DE 2175/2-1
RIKEN-CNRS, Collaborative Research Agreement
ANR, GRASP
CNRS, PEPS
CNRS, Neuro_IC2010
DAAD
LIA Vision for Action
Keywords Neuroscience
Electrophysiology
Utah Array
Spikes
Local Field Potential
Macaque
Motor Cortex
Resource Type Dataset