elephant-data/unittest/cell_assembly_detection

Cell assembly detection (CAD) test data

This folder contains the data used for CAD unit tests. Specifically, it uses the MATLAB [Registered Trademark] implementation of CAD from https://github.com/DurstewitzLab/Cell-Assembly-Detection .\ This implementation was used in: \ Eleonora Russo, Daniel Durstewitz (2017) Cell assemblies at multiple time scales with arbitrary lag constellations eLife 6:e19428, https://doi.org/10.7554/eLife.19428

Generating the validation dataset

The following steps will generate the validation data.

Run the MATLAB code for CAD

The validation data was generated using the MATLAB Code from GitHub DurstewitzLab/Cell-Assembly-Detection. The MATLAB code itself features a tutorial in Tutorial_main.m.\ Some test data is also provided in test_data.mat which contains a 50 by 7994 matrix.\ In this context the rows are interpreted as "units" and the columns contain the timestamps of the spikes.

To generate the data only the first 11 lines of the tutorial are executed, see Tutorial_main.m:

%%%%%%%%%%%%%%%%%%%%%%%%%% TUTORIAL ON CELL ASSEMBLY DETECTION %%%%%%%%%%%%%%%%%%%%%%%%%

% LOAD DATA
load('test_data.mat');
nneu=size(spM,1);  % nneu is number of recorded units

BinSizes=[0.015 0.025 0.04 0.06 0.085 0.15 0.25 0.4 0.6 0.85 1.5];
MaxLags=[10 10 10 10 10 10 10 10 10 10 10];

% ASSEMBY DETECTION
[assembly]=Main_assemblies_detection(spM,MaxLags,BinSizes);

with

• spM: the dataset loaded from test-data.mat
• nneu: number of recorded units (not used for this example)
• BinSizes: different bin sizes used for spiketrain binnning
• MaxLags: lags used for cell assembly detection

Executing the above lines of code will produce 11 result files: assembly1.mat ... assembly11.mat.

For the purpose of generating validation data, the naming for the files was changed to include the BinSize in the file name. Those changes were implemented in Main_assemblies_detection.m, filenames are assembly0015msec.mat ... assembly1500msec.mat.

The result files contain the bin_edges as well as n.

The MATLAB versions and toolboxes used to calculate the results were:

-----------------------------------------------------------------------------------------------------
MATLAB Version: 9.4.0.813654 (R2018a)
Operating System: Linux 5.15.0-56-generic #62-Ubuntu SMP Tue Nov 22 19:54:14 UTC 2022 x86_64
Java Version: Java 1.8.0_144-b01 with Oracle Corporation Java HotSpot(TM) 64-Bit Server VM mixed mode
-----------------------------------------------------------------------------------------------------
MATLAB                                                Version 9.4         (R2018a)
Signal Processing Toolbox                             Version 8.0         (R2018a)
Statistics and Machine Learning Toolbox               Version 11.3        (R2018a)

Contents

scripts/

• LICENSE (BSD 3-clause)
• authors -- list of authors and contributors *

data/

• LICENSE (CC-BY)
• test_data.mat which contains a 50 by 7994 matrix. In this context the rows are interpreted as "units" and the columns contain the timestamps of the spikes.

Copyright

:copyright: 2023 by the development team.

References and links

1. https://elifesciences.org/articles/19428
2. https://github.com/DurstewitzLab/Cell-Assembly-Detection