Computational toolbox for analysis of calcium imaging data of neuronal populations
Switch branches/tags
Nothing to show
Clone or download
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Toolbox software Fixed bug when 0 artifacts were found Nov 16, 2017
LICENSE Initial commit Feb 17, 2017
README.md Update on download of the test dataset Aug 10, 2017

README.md

Toolbox-Romano-et-al

Computational toolbox for analysis of calcium imaging data of neuronal populations

See the preprint A computational toolbox and step-by-step tutorial for the analysis of neuronal population dynamics in calcium imaging data by Romano et al for tutorial and full description (doi.org/10.1101/103879).

The toolbox is demonstrated in the paper An integrated calcium imaging processing toolbox for the analysis of neuronal population dynamics by Romano et al (doi.org/10.1371/journal.pcbi.1005526).

Developed at Zebrafish Neuroethology lab (http://www.zebrain.biologie.ens.fr/)

Installation

Download the latest version of the toolbox by clicking on "Clone or download" of this GitHub repository.

UPDATE August 2017: Due to the GitHub bandwidht limitation, the test dataset for the toolbox tutorial described in the biorxiv preprint has been moved here: http://www.zebrain.biologie.ens.fr/codes/ToolboxRomanoetal_TestData.zip When download completes, unzip and move the "Test data" folder to the the "Toolbox-Romano-et-al" folder downloaded from this GitHub repository. We apoplogize for problems experienced by users lately when trying to download the test dataset directly from GitHub. We were unaware of the bandwidth limitation on the Git LFS used to download the dataset from the repository. This limit was exceeded a while ago and Git LFS was automatically disabled for this project, making download of the dataset impossible, and we missed the warning email. We hope that by moving the test dataset we will avoid future problems.

Bug fixes

April 2017: Fixed bug in FindROIs.m that mistakenly stored inverted ROI perimeters (i.e., a "mirror image" of the correct ROIs).

Description of variables

We now describe all the variables stored in Matlab .mat files during the utilization of the toolbox. All the files correspond to processing and analysis of a fluorescence imaging video (e.g., myVideo.tif) with T imaging frames and N ROIs.

  • In the _ALL_CELLS.mat file (e.g., myVideo_ALL_CELLS.mat), the following variables are stored:

avg : average image (across frames) of the imaging file, showing the anatomy of imaged plane.

bkg : a logical mask containing all the ROIs found. Same size as avg.

cells : a Matlab cell array of size 1 x N, containing the pixel indexes of each ROI.

cell_number : total number of ROIs found (equal to N).

cells_mean : matrix of size T x N, containing the fluorescence time series of all the ROIs.

cell_per : a Matlab cell array of size N x 1, containing the perimeter coordinates for each ROI.

distances : matrix of size N x N, with the distances (in pixels) between ROIs.

npil_mean : matrix of size T x N, containing the fluorescence time series of the local perisomatic

neuropil that surrounds each ROIs.

pixelLengthX : size in micrometers of each image pixel in the X direction.

pixelLengthY : size in micrometers of each image pixel in the Y direction.

  • In the _ARTIFACTS.mat file (e.g., myVideo_ARTIFACTS.mat), the following variable is stored:

movements : T x 1 binary array, with ones for imaging frames where a movement artifact was found, and zeros otherwise.

  • In the _RASTER.mat file (e.g., myVideo_RASTER.mat), the following variables are stored:

raster : a T x N matrix of the significant fluorescence transients of all accepted ROIs.

deltaFoF : a T x N matrix of the ∆F/F0 time series for all the accepted ROIs.

F0 : a T x N matrix of baseline fluorescence time series matrix for all the accepted ROIs.

deletedCells : the indexes of the rejected ROIs.

movements : T x 1 binary array, with ones for imaging frames where a movement artifact was detected, zero otherwise. Same variable as that in _ARTIFACTS.mat file.

mu : 1 x N array with the average fluorescence baseline of each accepted ROI.

sigma : 1 x N array with the estimated fluorescence baseline noise scale.

mapOfOdds : map of fluorescence transitions that were considered significant.

mapOfOddsJoint : map of fluorescence transitions that were considered significant and biophysically realistic (rasters are determined with this map).

xev : x-axis of mapOfOddsJoint.

yev : y-axis of mapOfOddsJoint.

densityData : density of fluorescence transitions of all the accepted ROIs.

densityNoise : density of fluorescence transitions of the noise model.

imageAvg : the average image (across video frames) of your TIFF file.

params : parameters chosen by the user for the analysis of the fluorescence dynamics.

  • In the _RESPONSE_MAP.mat file (e.g., myVideo_RESPONSE_MAP.mat), the following variables are stored:

traces : Matlab cell array of size N x S, where S is the number of experimental events being mapped, where the ROI ∆F/F trial traces of each trial event are stored.

mapParams : S x 1 of the parametric values of the events.

roiHSV : N x 3 matrix with original HSV color-code for the mapped ROI responses.

roiHSVRescaled : same as roiHSV after color rescaling.

peakParameter : N x 1 matrix with the event parameter that gives a ROI peak response (mapped to hue color channel).

peakParameterStd : N x 1 matrix with the tuning width around the ROI peak response (mapped to the inverse of the saturation color channel).

peakResponse : N x 1 matrix with the ∆F/F value of the ROI peak response (mapped to the value color channel).

  • In the _CLUSTERS.mat file (e.g., myVideo_CLUSTERS.mat), the following variables are stored:

assembliesCells : Matlab cell array of size 1 x M, where M is the number of assemblies found, containing the cells that participate in each assembly.

confSynchBinary : threshold for significance of the pooled population activity count (events of synchronous population activity with cell count above this value are significant).

matchIndexTimeSeries : M x T matrix with the time series of the matching index for the assemblies.

matchIndexTimeSeriesSignificance : M x T matrix with the significance (p-Value) of the time series of the matching index for the assemblies.

threshSignifMatchIndex : threshold for significance used for the matching index time series of the assemblies.

matchIndexTimeSeriesSignificant : M x T matrix with the complete transients of significant assemblies' matching indexes.

matchIndexTimeSeriesSignificantPeaks : M x T matrix with the transient peaks of significant assemblies' matching indexes.

  • In the _ORDER_TOPO.mat file (e.g., myVideo_ORDER_TOPO.mat), the following variables are stored:

orderOfCells : order of all ROIs, according to the projection of the assemblies centroids over the anatomical axis drawn by the user (non-assembly cells are at the bottom).

orderOfCellsInAssemblies : same as orderOfCells, but only for assemblies' cells.

assembliesOrdered : Matlab cell array of size 1 x R, where R is the number of anatomical axes selected.

  • In the _SURROGATE_CLUSTERS.mat file (e.g., myVideo_SURROGATE_CLUSTERS.mat), the following variables are stored:

assembliesSurrogateRandom : Matlab cell array of size P x K, where K is the number of surrogate random assemblies per original assembly, with the list of ROIs in each random surrogate assembly.

assembliessSurrogateTopo : Matlab cell array of size P x K, where K is the number of surrogate topographic assemblies per original assembly, with the list of ROIs in each topographic surrogate assembly.

  • In the _ASSEMBLIES_vs_SURROGATE.mat file (e.g., myVideo_ASSEMBLIES_vs_SURROGATE.mat), the following variables are stored:

varAssemblies : Matlab cell array of size 1 x F, where F is the number of features. For feature Fi the variable varAssemblies{Fi}.data has the feature pooled according to the assemblies.

varSurrogate : Matlab cell array of size 1 x F, where F is the number of features. For feature Fi the variables varNMs{Fi}.Random and varNMs{Fi}.Topo have the feature pooled according to the random surrogate assemblies and the topographic surrogate assemblies, respectively.