cerebellum.analysis package

Module contents

Analysis and plotting tools for cerebellar cortex reconstructions.

class cerebellar_models.analysis.BSBReport(scaffold: str | Scaffold, cell_types_info: List[PlotTypeInfo] = None)

Bases: Report

Class interfacing a list of matplotlib plots for analysis of BSB reconstructions.

Parameters:
add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
property cell_names

List of the name of the elements to plot.

property labelled_abbreviations
property labelled_cell_names
class cerebellar_models.analysis.BasicSimulationReport(scaffold: str | Scaffold, simulation_name: str, folder_nio: str, time_from: float = 0, time_to: float = None, ignored_ct=None, cell_types_info: List[PlotTypeInfo] = None)

Bases: SpikeSimulationReport

Simulation report of the spike activity containing:

  • a plot with the raster and PSTH for each cell type,

  • a table plot storing the mean firing rate and ISI value for each cell type,

  • an instantaneous firing rate plot for each cell type,

  • an inter-spike interval histogram plot for each cell type,

  • a frequency spectrum plot for each cell type,

  • a legend plot

Parameters:
preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

class cerebellar_models.analysis.CellPlacement3D(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, ignored_ct=None, **kwargs)

Bases: ScaffoldPlot

Plot the position of the cells in the scaffold in 3D space.

Parameters:
ignored_ct

List of cell type names to ignore in the plot.

init_plot(**kwargs)

Initialize the plot and return figure and axes.

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

static set_axes_equal(ax)

Make axes of 3D plot have equal scale so that spheres appear as spheres, cubes as cubes, etc.

Parameters:

ax – a matplotlib axis, e.g., as output from plt.gca().

class cerebellar_models.analysis.ConnectivityTable(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, dict_abv=None, **kwargs)

Bases: TablePlot, ScaffoldPlot

Table plot of the results of the connectivity for BSB Scaffold. This includes for each pair of connected cell types:

  • the number of synapses formed

  • the number of synapses per unique pair of cell.

  • the convergence ratio defined as the mean number of afferent connections created with a single postsynaptic cell

  • the divergence ratio defined as the mean number of efferent connections created with a single presynaptic cell

Parameters:
dict_abv

Dictionary of abbreviations for cell types

extract_strat_name(ps: ConnectivitySet)

Convert the name of a connection into its abbreviation. Connection strategy name is assumed to be in the form A_B, where A is the presynaptic type name and B is the postsynaptic type name.

Parameters:

ps (ConnectivitySet)

get_convergences()

Return a dictionary which gives for each connection name the divergence number of each unique postsynaptic cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_divergences()

Return a dictionary which gives for each connection name the divergence number of each unique presynaptic cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_nb_synapse_per_pair()

Return a dictionary which gives for each connection name the number of synapses per unique pair of cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_synapse_counts()

Return a dictionary which gives for each connection name the number of synapses formed. The plot needs to be updated.

Return type:

Dict[str, int]

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.FiringRatesPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, kernel=None, dict_colors: dict = None, **kwargs)

Bases: Spike2Columns

Instantaneous firing rate plot for each cell type based on a time kernel. A firing rate signal is computed as the mean of the convolution of spike times for each neuron with the time kernel.

Parameters:
kernel

Elephant kernel to filter the spike trains

plot(relative_time=False, **kwargs)

Plot or replot the figure Calls the update function if needed.

Parameters:

relative_time (bool) – If True, the x-axis values will be relative to the time interval.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.FrequencyPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, kernel=None, dict_colors: dict = None, **kwargs)

Bases: FiringRatesPlot

Plot of the frequency distribution analysis of the instantaneous firing rate signal.

Parameters:
plot(max_freq=30.0, plot_bands=True, **kwargs)

Plot or replot the figure Calls the update function if needed.

Parameters:

  • max_freq (float) – maximum frequency (in Hz).

  • plot_bands (bool) – if True, plot the frequency bands.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.ISIPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 50, dict_colors: dict = None, **kwargs)

Bases: Spike2Columns

Inter-spike interval histogram plot for each cell type. For each neuron type, one mean inter-spike interval value is computed for each of its neuron.

Parameters:
nb_bins

Number of bins of the histogram.

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

class cerebellar_models.analysis.Legend(fig_size: Tuple[float, float], nb_cols: int = 1, dict_colors: dict = None, dict_abbreviations: dict = None, dict_legend=None, **kwargs)

Bases: Plot

Patch Legend plot

Parameters:
plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

remove_ct(to_keep: List[str], to_ignore: List[str] = None)

Remove cell types that are not in the list to keep or in the list to ignore

Params to_keep List[str]:

list of cell types to keep

Params to_ignore List[str]:

list of cell types to remove

Parameters:
class cerebellar_models.analysis.PlacementTable(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, dict_abv=None, **kwargs)

Bases: TablePlot, ScaffoldPlot

Table plot of the results of the placement for BSB Scaffold. This includes the counts and density of each cell type.

Parameters:
dict_abv

Dictionary of abbreviations for cell types

extract_ct_name(ct: CellType)

Convert the name of a cell type to its abbreviation.

Parameters:

ct (CellType)

get_counts()

Return a dictionary which gives for each cell type the number placed in the Scaffold. The plot needs to be updated.

Return type:

Dict[str, int]

get_densities()
get_volumes()

Return a dictionary which gives for each cell type the volume occupied by its cells in the Scaffold. The plot needs to be updated.

Return type:

Dict[str, int]

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.Plot(fig_size: Tuple[float, float] | ndarray, nb_rows: int = 1, nb_cols: int = 1, dict_colors=None, **kwargs)

Bases: ABC

Matplotlib plot class interface.

Parameters:
axes

Matplotlib Axes of the plot

clear()

Clear the figure axes

dict_colors

Dictionary of element name to their color

fig_size

Tuple size of the figure in inches.

figure

Matplotlib Figure of the plot.

get_ax(idx: int = 0)

Return the axis at the given index.

Parameters:

idx (int)

get_axes()

Return figure axes as a flat list.

init_plot(**kwargs)

Initialize the plot and return figure and axes.

is_initialized

Flag to indicate if this plot figure has been initialized.

is_plotted

Flag to indicate if this plot figure has been plotted.

is_updated

Flag to indicate if this plot data has been updated.

nb_cols

Number of sub-panel columns in the plot.

nb_rows

Number of sub-panel rows in the plot.

plot(*args, **kwargs)

Plot or replot the figure. Calls the update function if needed.

save_figure(output_name: str, dpi: int, pad=0, **kwargs)

Save the figure as a file.

Parameters:
set_axis_off(axes=None)

Removes the borders of the provided axes. If none are provided, all axes’ borders will be removed.

Parameters:

axes – List of matplotlib axes

set_color(key: str, color: ndarray[float])

Set the color dictionary for a given key. Colors must be an array of type RGB or RGBA.

Parameters:
show()

Show the figure. The figure will be plotted if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.PlotTypeInfo(name: str, color, abbreviation: str = None)

Bases: object

Class storing information about element plotted, usually cell types or fibers.

Parameters:

  • name (str)

  • abbreviation (str)

abbreviation

Abbreviation of the element to print in the plot.

color

Color of the element.

name

Name of the element.

class cerebellar_models.analysis.RasterPSTHPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 30, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Combined raster plot and PSTH plot of the spiking activity results for each neuron type. The subplots are split in two columns.

Parameters:
clear()

Clear the figure axes

init_plot(**kwargs)

Initialize the plot and return figure and axes.

nb_bins

Number of bins for the PSTH subplot.

plot(relative_time=False, params_raster: dict = None, params_psth: dict = None)

Plot or replot the figure Calls the update function if needed.

Parameters:

  • relative_time (bool) – If True, the x-axis values will be relative to the time interval.

  • params_raster (dict) – Dictionary of parameters for the raster plot (see matplotlib scatter).

  • params_psth (dict) – Dictionary of parameters for the PSTH plot (see matplotlib hist).

class cerebellar_models.analysis.Report(cell_types_info: List[PlotTypeInfo] = None)

Bases: object

Class interfacing a list of matplotlib plots to save in an external file.

Parameters:

cell_types_info (List[PlotTypeInfo])

property abbreviations

Dictionary from the name of the elements to plot to its abbreviation.

add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
cell_types_info

List of PlotTypeInfo for each element to plot.

property colors

Dictionary from the name of the elements to plot to its color.

plots

Dictionary mapping the report’s plots’ name to their instance

preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

print_report(output_name: str, dpi: int = 200, pad: int = 0, **kwargs)

Print the report and export it in a pdf file.

Parameters:
save_plot(plot_name: str, output_name: str, dpi: int = 200, pad: int = 0, **kwargs)

Save one of the report’s plots as a separate figure.

Parameters:
set_color(key: str, color: ndarray[float])

Set a color for an element to plot. Colors must be an array of type RGB or RGBA.

Parameters:
set_plot_colors(plot: Plot)

Set the plot color dictionary according to the report’s

Parameters:

plot (Plot)

show()

Show all report’s plots one by one.

show_plot(plot_name)

Show one of the report’s plots.

class cerebellar_models.analysis.RunStructureReport(*args, _parent=None, _key=None, **kwargs)

Bases: AfterConnectivityHook

BSB postprocessing node to generate a scaffold structural report after running the connectivity jobs.

get_node_name()
output_filename: str

Name of the pdf file to save the report.

postprocess()
class cerebellar_models.analysis.ScaffoldPlot(fig_size: Tuple[float, float], scaffold: Scaffold, nb_rows: int = 1, nb_cols: int = 1, dict_colors: dict = None, **kwargs)

Bases: Plot

Matplotlib plot interface for BSB Scaffold.

Parameters:
static get_labelled_ct_name(ct_name, labels)
static get_unique_labels(ps)

Get unique labels sets for a provided placement set Provide the default placement set label sets if the placement set is empty.

property labelled_dict_colors
set_scaffold(scaffold)

Set the scaffold of the plot.

class cerebellar_models.analysis.SimResultsTable(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, dict_abv=None, **kwargs)

Bases: TablePlot, SpikePlot

Table of the firing rates and inter-spike intervals for each cell type. The firing rate value of a cell type corresponds to the mean number of spike over the time interval, while its inter-spike interval corresponds to the mean of all mean inter-spike interval values computed for each of its neuron.

Parameters:
dict_abv

Dictionary of abbreviations for cell types

get_firing_rates()

Return a dictionary which gives for each cell type the firing rate of each neuron spiking. The plot needs to be updated.

Return type:

Dict[str, int]

get_isis_values()
plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.SortedPSTH(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 50, sample_size: int = 100, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Parameters:
nb_bins

Number of bins for the PSTH.

plot()

Plot or replot the figure. Calls the update function if needed.

sample_size

Maximum number of neurons to subsample for each population.

class cerebellar_models.analysis.Spike2Columns(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Utility class to plot simulation results for each neuron type in a 2 columns fashion.

Parameters:
init_plot(**kwargs)

Initialize the plot and return figure and axes.

class cerebellar_models.analysis.SpikeCorrelationPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, bin_size: float = array(5.) * ms, dict_colors: dict = None, dict_abv=None, **kwargs)

Bases: SpikePlot

Spike cross-correlation matrix plot for each cell type. Spike trains will be time binned before computing the pairwise Pearson’s correlation coefficients.

Parameters:
bin_size

Size of the time bins used to group spikes before computing correlation coefficients.

dict_abv

Dictionary of abbreviations for cell types

plot()

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.SpikePlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, **kwargs)

Bases: ScaffoldPlot

Abstract class for plotting the spiking simulation results of a BSB scaffold.

Parameters:
property dt
property filt_spikes
property nb_neurons
property populations
spiking_results

Loaded spiking results.

property time_from
property time_to
class cerebellar_models.analysis.SpikeSimulationReport(scaffold: str | Scaffold, simulation_name: str, folder_nio: str, time_from: float = 0, time_to: float = None, ignored_ct=None, cell_types_info: List[PlotTypeInfo] = None)

Bases: BSBReport

Abstract class for reports of simulation results of BSB scaffold.

Parameters:
add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
property dt
property filt_spikes

Filter the spike events for the time of the analysis.

Returns:

Sliced List of SpikeTrain.

Return type:

List[neo.core.SpikeTrain]

property nb_neurons
property populations
property time_from
property time_to
class cerebellar_models.analysis.SpikingResults(scaffold: Scaffold, simulation_name, time_from: float, time_to: float, folder_nio: str, ignored_ct: list[str])

Bases: object

Class used to load the results of the NEST simulations produced from a BSB scaffold.

Parameters:

  • scaffold (Scaffold) – BSB scaffold

  • simulation_name – name of the simulation

  • time_from (float) – start time of the analysis

  • time_to (float) – end time of the analysis

  • folder_nio (str) – folder where the Neo results are stored

  • ignored_ct (list[str]) – List of ignored cells names

property dt

Time step of the simulation in ms

property filt_spikes: List[SpikeTrain]

Filter the spike events for the time of the analysis.

Returns:

List of time-sliced SpikeTrain.

Return type:

List[neo.core.SpikeTrain]

property folder_nio
ignored_ct

List of ignored cell type names

load_spikes()

Load the spike trains from nio files.

Returns:

  • Boolean numpy array of shape (N*M) storing spike events for each time step. N corresponds to the number of time steps, M to the number of neuron. Neurons are sorted by type.

  • List of number of unique neuron per type.

  • List of cell type names.

Return type:

Tuple[List[neo.core.SpikeTrain], numpy.ndarray[int], List[str]]

property nb_neurons: ndarray

Number of neuron for each neuron type

property populations: List[str]

List of neuron type names

property scaffold

BSB Scaffold used as reference for simulation results.

property simulation_name

Name of the simulation as defined in the scaffold configuration.

property time_from

Start time of the analysis

property time_to

End time of the analysis. By default, this corresponds to the simulation duration.

class cerebellar_models.analysis.StructureReport(scaffold: str | Scaffold, cell_type_info: List[PlotTypeInfo] = None)

Bases: BSBReport

Report of the scaffold neural network structure containing:

  • a table listing the number and density of each cell type placed.

  • a table listing the number, convergence and divergence of connections for each connected cell type pair

  • a 3D plot showing the soma location of each cell in the circuit.

  • a legend plot

Parameters:
preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

class cerebellar_models.analysis.TablePlot

Bases: object

Mixin for plotting tables with matplotlib

columns = []

Names of the table’s columns

plot_table(**kwargs)

Plot the table in the Figure.

reset_table()

Update the values of the table.

rows = []

Names of the table’s rows

table_values = []

List of list of string values to put in the table

cerebellar_models.analysis.extract_isis(spikes, dt)

Extract inter-spike intervals from a SpikeTrain object. One mean inter-spike interval value is computed for each neuron.

Parameters:
Returns:

list of inter-spike intervals

Return type:

List[float]

cerebellar_models.analysis.get_correlation_coefficients(spiking_results, bin_size)

Get the spike cross-correlation matrix for each cell type. Spike trains will be time binned before computing the pairwise Pearson’s correlation coefficients.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • bin_size (float) – size of time bin

Returns:

numpy array 2D matrix storing the Pearson correlation coefficients between each neuron population.

Return type:

numpy.ndarray[float]

cerebellar_models.analysis.get_firing_rates(spiking_results: SpikingResults, kernel=None) → ndarray

Get the instantaneous firing rate for each cell type from SpikingResults based on a time kernel.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • kernel – Elephant kernel to filter the spike trains

Returns:

numpy array storing instantaneous firing rates for each population, for each time step.

Return type:

numpy.ndarray[float]

cerebellar_models.analysis.get_frequencies(spiking_results, firing_rates)

Get the Fast Fourier Transform on instantaneous firing rates signal.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • firing_rates – numpy array storing instantaneous firing rates for each population, for each time step.

Returns:

Tuple of lists of frequencies and their corresponding FFT powers.

Return type:

Tuple[numpy.ndarray[float], numpy.ndarray[float]]

Submodules

cerebellar_models.analysis.plots module

Module for abstract classes interfacing matplotlib plots.

class cerebellar_models.analysis.plots.Legend(fig_size: Tuple[float, float], nb_cols: int = 1, dict_colors: dict = None, dict_abbreviations: dict = None, dict_legend=None, **kwargs)

Bases: Plot

Patch Legend plot

Parameters:
plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

remove_ct(to_keep: List[str], to_ignore: List[str] = None)

Remove cell types that are not in the list to keep or in the list to ignore

Params to_keep List[str]:

list of cell types to keep

Params to_ignore List[str]:

list of cell types to remove

Parameters:
class cerebellar_models.analysis.plots.Plot(fig_size: Tuple[float, float] | ndarray, nb_rows: int = 1, nb_cols: int = 1, dict_colors=None, **kwargs)

Bases: ABC

Matplotlib plot class interface.

Parameters:
axes

Matplotlib Axes of the plot

clear()

Clear the figure axes

dict_colors

Dictionary of element name to their color

fig_size

Tuple size of the figure in inches.

figure

Matplotlib Figure of the plot.

get_ax(idx: int = 0)

Return the axis at the given index.

Parameters:

idx (int)

get_axes()

Return figure axes as a flat list.

init_plot(**kwargs)

Initialize the plot and return figure and axes.

is_initialized

Flag to indicate if this plot figure has been initialized.

is_plotted

Flag to indicate if this plot figure has been plotted.

is_updated

Flag to indicate if this plot data has been updated.

nb_cols

Number of sub-panel columns in the plot.

nb_rows

Number of sub-panel rows in the plot.

plot(*args, **kwargs)

Plot or replot the figure. Calls the update function if needed.

save_figure(output_name: str, dpi: int, pad=0, **kwargs)

Save the figure as a file.

Parameters:
set_axis_off(axes=None)

Removes the borders of the provided axes. If none are provided, all axes’ borders will be removed.

Parameters:

axes – List of matplotlib axes

set_color(key: str, color: ndarray[float])

Set the color dictionary for a given key. Colors must be an array of type RGB or RGBA.

Parameters:
show()

Show the figure. The figure will be plotted if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.plots.ScaffoldPlot(fig_size: Tuple[float, float], scaffold: Scaffold, nb_rows: int = 1, nb_cols: int = 1, dict_colors: dict = None, **kwargs)

Bases: Plot

Matplotlib plot interface for BSB Scaffold.

Parameters:
static get_labelled_ct_name(ct_name, labels)
static get_unique_labels(ps)

Get unique labels sets for a provided placement set Provide the default placement set label sets if the placement set is empty.

property labelled_dict_colors
set_scaffold(scaffold)

Set the scaffold of the plot.

cerebellar_models.analysis.report module

Module for the Report class.

class cerebellar_models.analysis.report.BSBReport(scaffold: str | Scaffold, cell_types_info: List[PlotTypeInfo] = None)

Bases: Report

Class interfacing a list of matplotlib plots for analysis of BSB reconstructions.

Parameters:
add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
property cell_names

List of the name of the elements to plot.

property labelled_abbreviations
property labelled_cell_names
cerebellar_models.analysis.report.LIST_CT_INFO = [<cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>, <cerebellar_models.analysis.report.PlotTypeInfo object>]

Cell and fiber information for plotting cerebellum BSB circuit.

class cerebellar_models.analysis.report.PlotTypeInfo(name: str, color, abbreviation: str = None)

Bases: object

Class storing information about element plotted, usually cell types or fibers.

Parameters:

  • name (str)

  • abbreviation (str)

abbreviation

Abbreviation of the element to print in the plot.

color

Color of the element.

name

Name of the element.

class cerebellar_models.analysis.report.Report(cell_types_info: List[PlotTypeInfo] = None)

Bases: object

Class interfacing a list of matplotlib plots to save in an external file.

Parameters:

cell_types_info (List[PlotTypeInfo])

property abbreviations

Dictionary from the name of the elements to plot to its abbreviation.

add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
cell_types_info

List of PlotTypeInfo for each element to plot.

property colors

Dictionary from the name of the elements to plot to its color.

plots

Dictionary mapping the report’s plots’ name to their instance

preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

print_report(output_name: str, dpi: int = 200, pad: int = 0, **kwargs)

Print the report and export it in a pdf file.

Parameters:
save_plot(plot_name: str, output_name: str, dpi: int = 200, pad: int = 0, **kwargs)

Save one of the report’s plots as a separate figure.

Parameters:
set_color(key: str, color: ndarray[float])

Set a color for an element to plot. Colors must be an array of type RGB or RGBA.

Parameters:
set_plot_colors(plot: Plot)

Set the plot color dictionary according to the report’s

Parameters:

plot (Plot)

show()

Show all report’s plots one by one.

show_plot(plot_name)

Show one of the report’s plots.

cerebellar_models.analysis.structure_analysis module

Module for the plots and reports related to the structural analysis of BSB scaffold.

class cerebellar_models.analysis.structure_analysis.AdjacencyMatrix(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, ignored_ct=None, dict_abv=None, **kwargs)

Bases: ScaffoldPlot

Plot showing the cell-type adjacency matrix of a scaffold.

Parameters:
adjacency_matrix

Cell to cell adjacency matrix.

dict_abv

Dictionary of abbreviations for cell types

grouped_matrix

Cell type population to cell type population adjacency matrix.

ignored_ct

List of cell type names to ignore in the plot.

normalize()

Normalization function for resulting the adjacency matrix.

plot(log_scale=True, **kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.structure_analysis.CellPlacement3D(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, ignored_ct=None, **kwargs)

Bases: ScaffoldPlot

Plot the position of the cells in the scaffold in 3D space.

Parameters:
ignored_ct

List of cell type names to ignore in the plot.

init_plot(**kwargs)

Initialize the plot and return figure and axes.

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

static set_axes_equal(ax)

Make axes of 3D plot have equal scale so that spheres appear as spheres, cubes as cubes, etc.

Parameters:

ax – a matplotlib axis, e.g., as output from plt.gca().

class cerebellar_models.analysis.structure_analysis.ConnectivityTable(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, dict_abv=None, **kwargs)

Bases: TablePlot, ScaffoldPlot

Table plot of the results of the connectivity for BSB Scaffold. This includes for each pair of connected cell types:

  • the number of synapses formed

  • the number of synapses per unique pair of cell.

  • the convergence ratio defined as the mean number of afferent connections created with a single postsynaptic cell

  • the divergence ratio defined as the mean number of efferent connections created with a single presynaptic cell

Parameters:
dict_abv

Dictionary of abbreviations for cell types

extract_strat_name(ps: ConnectivitySet)

Convert the name of a connection into its abbreviation. Connection strategy name is assumed to be in the form A_B, where A is the presynaptic type name and B is the postsynaptic type name.

Parameters:

ps (ConnectivitySet)

get_convergences()

Return a dictionary which gives for each connection name the divergence number of each unique postsynaptic cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_divergences()

Return a dictionary which gives for each connection name the divergence number of each unique presynaptic cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_nb_synapse_per_pair()

Return a dictionary which gives for each connection name the number of synapses per unique pair of cell. The plot needs to be updated.

Return type:

Dict[str, numpy.ndarray[int]]

get_synapse_counts()

Return a dictionary which gives for each connection name the number of synapses formed. The plot needs to be updated.

Return type:

Dict[str, int]

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.structure_analysis.PlacementTable(fig_size: Tuple[float, float], scaffold: Scaffold = None, dict_colors=None, dict_abv=None, **kwargs)

Bases: TablePlot, ScaffoldPlot

Table plot of the results of the placement for BSB Scaffold. This includes the counts and density of each cell type.

Parameters:
dict_abv

Dictionary of abbreviations for cell types

extract_ct_name(ct: CellType)

Convert the name of a cell type to its abbreviation.

Parameters:

ct (CellType)

get_counts()

Return a dictionary which gives for each cell type the number placed in the Scaffold. The plot needs to be updated.

Return type:

Dict[str, int]

get_densities()
get_volumes()

Return a dictionary which gives for each cell type the volume occupied by its cells in the Scaffold. The plot needs to be updated.

Return type:

Dict[str, int]

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.structure_analysis.RunStructureReport(*args, _parent=None, _key=None, **kwargs)

Bases: AfterConnectivityHook

BSB postprocessing node to generate a scaffold structural report after running the connectivity jobs.

get_node_name()
output_filename: str

Name of the pdf file to save the report.

postprocess()
class cerebellar_models.analysis.structure_analysis.StructureReport(scaffold: str | Scaffold, cell_type_info: List[PlotTypeInfo] = None)

Bases: BSBReport

Report of the scaffold neural network structure containing:

  • a table listing the number and density of each cell type placed.

  • a table listing the number, convergence and divergence of connections for each connected cell type pair

  • a 3D plot showing the soma location of each cell in the circuit.

  • a legend plot

Parameters:
preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

class cerebellar_models.analysis.structure_analysis.TablePlot

Bases: object

Mixin for plotting tables with matplotlib

columns = []

Names of the table’s columns

plot_table(**kwargs)

Plot the table in the Figure.

reset_table()

Update the values of the table.

rows = []

Names of the table’s rows

table_values = []

List of list of string values to put in the table

cerebellar_models.analysis.spike_plots module

Module for the plots and reports related to the simulation analysis of BSB scaffold.

class cerebellar_models.analysis.spike_plots.BasicSimulationReport(scaffold: str | Scaffold, simulation_name: str, folder_nio: str, time_from: float = 0, time_to: float = None, ignored_ct=None, cell_types_info: List[PlotTypeInfo] = None)

Bases: SpikeSimulationReport

Simulation report of the spike activity containing:

  • a plot with the raster and PSTH for each cell type,

  • a table plot storing the mean firing rate and ISI value for each cell type,

  • an instantaneous firing rate plot for each cell type,

  • an inter-spike interval histogram plot for each cell type,

  • a frequency spectrum plot for each cell type,

  • a legend plot

Parameters:
preprocessing()

Function to apply modifications to the report’s plots after the plotting is done.

class cerebellar_models.analysis.spike_plots.FiringRatesPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, kernel=None, dict_colors: dict = None, **kwargs)

Bases: Spike2Columns

Instantaneous firing rate plot for each cell type based on a time kernel. A firing rate signal is computed as the mean of the convolution of spike times for each neuron with the time kernel.

Parameters:
kernel

Elephant kernel to filter the spike trains

plot(relative_time=False, **kwargs)

Plot or replot the figure Calls the update function if needed.

Parameters:

relative_time (bool) – If True, the x-axis values will be relative to the time interval.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.spike_plots.FrequencyPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, kernel=None, dict_colors: dict = None, **kwargs)

Bases: FiringRatesPlot

Plot of the frequency distribution analysis of the instantaneous firing rate signal.

Parameters:
plot(max_freq=30.0, plot_bands=True, **kwargs)

Plot or replot the figure Calls the update function if needed.

Parameters:

  • max_freq (float) – maximum frequency (in Hz).

  • plot_bands (bool) – if True, plot the frequency bands.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.spike_plots.ISIPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 50, dict_colors: dict = None, **kwargs)

Bases: Spike2Columns

Inter-spike interval histogram plot for each cell type. For each neuron type, one mean inter-spike interval value is computed for each of its neuron.

Parameters:
nb_bins

Number of bins of the histogram.

plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

class cerebellar_models.analysis.spike_plots.RasterPSTHPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 30, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Combined raster plot and PSTH plot of the spiking activity results for each neuron type. The subplots are split in two columns.

Parameters:
clear()

Clear the figure axes

init_plot(**kwargs)

Initialize the plot and return figure and axes.

nb_bins

Number of bins for the PSTH subplot.

plot(relative_time=False, params_raster: dict = None, params_psth: dict = None)

Plot or replot the figure Calls the update function if needed.

Parameters:

  • relative_time (bool) – If True, the x-axis values will be relative to the time interval.

  • params_raster (dict) – Dictionary of parameters for the raster plot (see matplotlib scatter).

  • params_psth (dict) – Dictionary of parameters for the PSTH plot (see matplotlib hist).

class cerebellar_models.analysis.spike_plots.SimResultsTable(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, dict_abv=None, **kwargs)

Bases: TablePlot, SpikePlot

Table of the firing rates and inter-spike intervals for each cell type. The firing rate value of a cell type corresponds to the mean number of spike over the time interval, while its inter-spike interval corresponds to the mean of all mean inter-spike interval values computed for each of its neuron.

Parameters:
dict_abv

Dictionary of abbreviations for cell types

get_firing_rates()

Return a dictionary which gives for each cell type the firing rate of each neuron spiking. The plot needs to be updated.

Return type:

Dict[str, int]

get_isis_values()
plot(**kwargs)

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.spike_plots.SortedPSTH(fig_size: Tuple[float, float], spiking_results: SpikingResults, nb_bins: int = 50, sample_size: int = 100, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Parameters:
nb_bins

Number of bins for the PSTH.

plot()

Plot or replot the figure. Calls the update function if needed.

sample_size

Maximum number of neurons to subsample for each population.

class cerebellar_models.analysis.spike_plots.Spike2Columns(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, **kwargs)

Bases: SpikePlot

Utility class to plot simulation results for each neuron type in a 2 columns fashion.

Parameters:
init_plot(**kwargs)

Initialize the plot and return figure and axes.

class cerebellar_models.analysis.spike_plots.SpikeCorrelationPlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, bin_size: float = array(5.) * ms, dict_colors: dict = None, dict_abv=None, **kwargs)

Bases: SpikePlot

Spike cross-correlation matrix plot for each cell type. Spike trains will be time binned before computing the pairwise Pearson’s correlation coefficients.

Parameters:
bin_size

Size of the time bins used to group spikes before computing correlation coefficients.

dict_abv

Dictionary of abbreviations for cell types

plot()

Plot or replot the figure. Calls the update function if needed.

update()

Update function to prepare the data before plotting.

class cerebellar_models.analysis.spike_plots.SpikePlot(fig_size: Tuple[float, float], spiking_results: SpikingResults, dict_colors: dict = None, **kwargs)

Bases: ScaffoldPlot

Abstract class for plotting the spiking simulation results of a BSB scaffold.

Parameters:
property dt
property filt_spikes
property nb_neurons
property populations
spiking_results

Loaded spiking results.

property time_from
property time_to
class cerebellar_models.analysis.spike_plots.SpikeSimulationReport(scaffold: str | Scaffold, simulation_name: str, folder_nio: str, time_from: float = 0, time_to: float = None, ignored_ct=None, cell_types_info: List[PlotTypeInfo] = None)

Bases: BSBReport

Abstract class for reports of simulation results of BSB scaffold.

Parameters:
add_plot(name: str, plot: Plot)

Add a plot to the list of the report’s plots.

Parameters:
property dt
property filt_spikes

Filter the spike events for the time of the analysis.

Returns:

Sliced List of SpikeTrain.

Return type:

List[neo.core.SpikeTrain]

property nb_neurons
property populations
property time_from
property time_to

cerebellar_models.analysis.spiking_results module

Module for loading spiking results from BSB-NEST simulations and analysing them with Elephant.

class cerebellar_models.analysis.spiking_results.SpikingResults(scaffold: Scaffold, simulation_name, time_from: float, time_to: float, folder_nio: str, ignored_ct: list[str])

Bases: object

Class used to load the results of the NEST simulations produced from a BSB scaffold.

Parameters:

  • scaffold (Scaffold) – BSB scaffold

  • simulation_name – name of the simulation

  • time_from (float) – start time of the analysis

  • time_to (float) – end time of the analysis

  • folder_nio (str) – folder where the Neo results are stored

  • ignored_ct (list[str]) – List of ignored cells names

property dt

Time step of the simulation in ms

property filt_spikes: List[SpikeTrain]

Filter the spike events for the time of the analysis.

Returns:

List of time-sliced SpikeTrain.

Return type:

List[neo.core.SpikeTrain]

property folder_nio
ignored_ct

List of ignored cell type names

load_spikes()

Load the spike trains from nio files.

Returns:

  • Boolean numpy array of shape (N*M) storing spike events for each time step. N corresponds to the number of time steps, M to the number of neuron. Neurons are sorted by type.

  • List of number of unique neuron per type.

  • List of cell type names.

Return type:

Tuple[List[neo.core.SpikeTrain], numpy.ndarray[int], List[str]]

property nb_neurons: ndarray

Number of neuron for each neuron type

property populations: List[str]

List of neuron type names

property scaffold

BSB Scaffold used as reference for simulation results.

property simulation_name

Name of the simulation as defined in the scaffold configuration.

property time_from

Start time of the analysis

property time_to

End time of the analysis. By default, this corresponds to the simulation duration.

cerebellar_models.analysis.spiking_results.extract_isis(spikes, dt)

Extract inter-spike intervals from a SpikeTrain object. One mean inter-spike interval value is computed for each neuron.

Parameters:
Returns:

list of inter-spike intervals

Return type:

List[float]

cerebellar_models.analysis.spiking_results.get_correlation_coefficients(spiking_results, bin_size)

Get the spike cross-correlation matrix for each cell type. Spike trains will be time binned before computing the pairwise Pearson’s correlation coefficients.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • bin_size (float) – size of time bin

Returns:

numpy array 2D matrix storing the Pearson correlation coefficients between each neuron population.

Return type:

numpy.ndarray[float]

cerebellar_models.analysis.spiking_results.get_firing_rates(spiking_results: SpikingResults, kernel=None) → ndarray

Get the instantaneous firing rate for each cell type from SpikingResults based on a time kernel.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • kernel – Elephant kernel to filter the spike trains

Returns:

numpy array storing instantaneous firing rates for each population, for each time step.

Return type:

numpy.ndarray[float]

cerebellar_models.analysis.spiking_results.get_frequencies(spiking_results, firing_rates)

Get the Fast Fourier Transform on instantaneous firing rates signal.

Parameters:

  • spiking_results (SpikingResults) – simulation spike results

  • firing_rates – numpy array storing instantaneous firing rates for each population, for each time step.

Returns:

Tuple of lists of frequencies and their corresponding FFT powers.

Return type:

Tuple[numpy.ndarray[float], numpy.ndarray[float]]

cerebellar_models.analysis.spiking_results.get_spike_matrix(spikes, dt)

Extract the 2D boolean matrix of the spiking activity for each neuron in the SpikeTrain object. Neurons are sorted according to their NEST id.

Parameters:
Returns:

numpy array 2D boolean matrix storing spike events for each neuron, for each time step.

Return type:

numpy.ndarray[bool]