griottes package

graphmaker

Tools for graph generation

graph_generation_func

griottes.graphmaker.graph_generation_func.generate_geometric_graph(user_entry, descriptors: list = [], distance: float = 60, image_is_2D=False, min_area=0, analyze_fluo_channels=False, radius=30, mask_channel=None)

Creates a geometric graph. This function creates a geometric graph from an image or a dataframe object. :param user_entry: contains the information on the cells. :type user_entry: pandas.DataFrame or numpy.ndarray :param descriptors: contains the cell information included in the

network nodes.

Parameters:

distance (float, optional) – the maximum distance between two nodes.
image_is_2D (bool, optional) – if True, the image is analyzed as a 2D image. The default is False.
min_area (int, optional) – the minimum area of a cell. The default is 0.
analyze_fluo_channels (bool, optional) – if True, the fluorescence channels are analyzed. The default is False.
radius (int, optional) – Radius of the sphere within the which the fluorescence is analyzed. Irrelevant for the ‘basic’ method. The default is 30.
mask_channel (int, optional) – The channel containing the cell masks The default is None.

Returns:

The graph representation of the input.

Return type:

nx.Graph

griottes.graphmaker.graph_generation_func.generate_contact_graph(labels_array: ndarray, dataframe_with_descriptors: Optional[DataFrame] = None, mask_channel=None, min_area=0, analyze_fluo_channels=True, image_is_2D=True, fluo_channel_analysis_method='basic', descriptors=[], radius=30, **kwargs)

Creates a contact graph. This function creates a contact graph from an image. The contact graph is a graph where each node represents a region and each edge represents a contact between two adjascent regions. :param labels_array: contains the information on the cells. :type labels_array: numpy.ndarray :param dataframe_with_descriptors: Labels, coordinates and, you know, whatever you want to be included in the graph :type dataframe_with_descriptors: pandas.DataFrame :param descriptors: contains the cell information included in the

network nodes.

Parameters:

image_is_2D (bool, optional) – if True, the image is analyzed as a 2D image. The default is False.
min_area (int, optional) – the minimum area of a cell. The default is 0.
analyze_fluo_channels (bool, optional) – if True, the fluorescence channels are analyzed. The default is True.
radius (int, optional) – Radius of the sphere within the which the fluorescence is analyzed. Irrelevant for the ‘basic’ method. The default is 30.
mask_channel (int, optional) – The channel containing the cell masks The default is None.

Returns:

The graph representation of the input.

Return type:

nx.Graph

griottes.graphmaker.graph_generation_func.generate_delaunay_graph(user_entry, descriptors: list = [], image_is_2D=False, min_area=0, analyze_fluo_channels=False, fluo_channel_analysis_method='basic', radius=30, distance=30, mask_channel=None)

Creates a Delaunay graph. This function creates a Delaunay graph from an image or a dataframe object. :param user_entry: contains the information on the cells. :type user_entry: pandas.DataFrame or numpy.ndarray :param descriptors: contains the cell information included in the

network nodes.

Parameters:

distance (float, optional) – the maximum distance between two nodes.
fluo_channel_analysis_method (str, optional) – the method used to analyze the fluorescence channels. ‘basic’ measures the fluorescence properties within the cell mask, ‘local_sphere’ within a sphere of radius ‘radius’ and ‘local_voronoi’ within the Voronoi tesselation of the cell.
radius (float, optional) – radius of the sphere within the which the fluorescence is analyzed. Irrelevant for the ‘basic’ fluorescence analysis method.
image_is_2D (bool, optional) – if True, the image is analyzed as a 2D image. The default is False.
min_area (int, optional) – the minimum area of a cell. The default is 0.
analyze_fluo_channels (bool, optional) – if True, the fluorescence channels are analyzed. The default is False.
mask_channel (int, optional) – The channel containing the cell masks The default is None.

Returns:

The graph representation of the input.

Return type:

nx.Graph

griottes.graphmaker.graph_generation_func.trim_graph_voronoi(G, distance, image_is_2D)

Remove slinks above the distance length. Serves to remove unrealistic edges from the graph. :param G: The graph representation of the input image/table. :type G: nx.Graph :param distance: The maximum distance between two nodes. :type distance: float :param image_is_2D: If True, the image is 2D. :type image_is_2D: bool

Returns:: The graph representation of the input.
Return type:: nx.Graph

griottes.graphmaker.graph_generation_func.get_region_contacts_2D(mask_image): From the masked image create a dataframe containing the information on all the links between region.

griottes.graphmaker.graph_generation_func.get_region_contacts_3D(mask_image): From the masked image create a dataframe containing the information on all the links between region.

make_spheroids

griottes.graphmaker.make_spheroids.single_spheroid_process(spheroid_frame: DataFrame, descriptors: list = []): spheroid_frame has the label of each cell as index.

graph_plot

griottes.graphplotter.graph_plot.network_plot_2D(G, background_image=None, figsize: tuple = (8, 8), alpha_line=0.6, scatterpoint_size=20, legend=False, edge_color='k', line_factor=1, legend_fontsize=18, include_weights=False)

griottes.graphplotter.graph_plot.network_plot_3D(G, figsize: tuple = (8, 8), alpha_line=0.6, scatterpoint_size=20, legend=False, legend_fontsize=12, theta=0, psi=0, xlim=None, ylim=None, zlim=None)

griottes.graphplotter.graph_plot.Voronoi_3D(G, angle)

analyse

cell_property_extraction

griottes.analyse.cell_property_extraction.get_nuclei_properties(image, mask_channel)

Get properties of nuclei in image.

Parameters:

image (numpy.ndarray) – Image with nuclei masks.
mask_channel (int) – Channel of the mask.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.get_shape_properties(properties, image, mask_channel, min_area, ndim)

Calculate shape properties of the nuclei.

Parameters:

properties (pandas.DataFrame) – Dataframe containing centroid, area, and label of the nuclei.
image (numpy.ndarray) – Image with nuclei masks.
mask_channel (int) – Channel of the mask.
min_area (int) – Minimum area of the nuclei to include in analysis.
ndim (int) – Number of dimensions in the image (2 or 3).

Returns:

Dataframe with additional columns for shape properties.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.get_fluo_properties(image, fluo_channel, mask_channel=0)

Get mean fluorescence intensity of each labeled region in a mask.

Parameters:

image (numpy.ndarray) – Multichannel image containing the fluorescence channel.
fluo_channel (int) – Channel containing the fluorescence signal.
mask_channel (int, optional) – Channel containing the mask. Default is 0.

Returns:

Dataframe containing the mean fluorescence intensity and label of each region.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.basic_fluo_prop_analysis(properties, image, mask_channel)

Calculate the mean intensity of each fluorescence channel for each nucleus.

Parameters:

properties (pandas.DataFrame) – DataFrame containing properties of each nucleus.
image (numpy.ndarray) – Image with fluorescence channels.
mask_channel (int) – Channel of the mask.

Returns:

Updated properties DataFrame with mean fluorescence intensity for each channel.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.sphere_mean_intensity(intensity_image, position, radius, percentile=50)

Calculate the mean intensity and percentile of a sphere within an intensity image.

Parameters:

intensity_image (numpy.ndarray) – The 3D intensity image.
position (tuple) – The center of the sphere in (z, y, x) format.
radius (float) – The radius of the sphere.
percentile (float, optional) – The percentile to calculate, default is 50.

Returns:

The mean intensity and percentile.

Return type:

tuple of float

griottes.analyse.cell_property_extraction.calculate_sphere_fluo_properties(properties: DataFrame, image: ndarray, fluo_channel: int, radius: int, mask_channel: int, percentile: int) → DataFrame

Calculate mean and percentile fluorescence intensity within a sphere around each nucleus.

Parameters:

properties (pandas.DataFrame) – Properties of the nuclei.
image (numpy.ndarray) – Image with fluorescence channels.
fluo_channel (int) – Channel containing the fluorescence signal.
radius (int) – Radius of the sphere around each nucleus.
mask_channel (int) – Channel of the nuclei mask.
percentile (int) – Percentile of the fluorescence intensity to calculate.

Returns:

Mean and percentile fluorescence intensity within the sphere around each nucleus.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.sphere_fluo_property_analysis(properties, image, mask_channel, radius, percentile)

Calculate sphere-based fluorescence properties for each label in the mask.

Parameters:

properties (pandas.DataFrame) – DataFrame with properties of each label in the mask.
image (numpy.ndarray) – 3D image stack.
mask_channel (int) – Index of the channel containing the mask.
radius (int) – Radius of the sphere used to calculate properties.
percentile (int) – Percentile used to calculate the fluorescence intensity.

Returns:

Updated DataFrame with fluorescence properties for each label.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.in_hull(p, hull)

Test if points in p are in hull

p should be a NxK coordinates of N points in K dimensions hull is either a scipy.spatial.Delaunay object or the MxK array of the coordinates of M points in `K`dimensions for which Delaunay triangulation will be computed

griottes.analyse.cell_property_extraction.make_spherical_mask(image, point_coordinates, radius)

Create a spherical mask centered at point_coordinates with radius.

Parameters:

image (numpy.ndarray) – Input image.
point_coordinates (tuple of int) – Coordinates of the center of the sphere in the format (z, x, y).
radius (int) – Radius of the sphere.

Returns:

A binary mask of the same shape as image, where voxels inside the sphere are set to 1 and voxels outside the sphere are set to 0.

Return type:

numpy.ndarray

griottes.analyse.cell_property_extraction.make_voronoi_mask(properties, image, mask_channel, radius)

Create a Voronoi diagram of the nuclei centroids and use it to generate a binary mask.

Parameters:

properties (pandas.DataFrame) – DataFrame with properties of the nuclei.
image (numpy.ndarray) – Image with nuclei masks.
mask_channel (int) – Channel of the mask.
radius (float) – Radius of the spherical mask used to generate the Voronoi diagram.

Returns:

Binary mask with the same shape as the input image, where each pixel is assigned to the nearest nucleus based on the Voronoi diagram.

Return type:

numpy.ndarray

griottes.analyse.cell_property_extraction.get_fluo_properties_voronoi(properties, image, fluo_channel, label_matrix, percentile)

Get fluorescence properties for each cell in a Voronoi diagram.

Parameters:

properties (pandas.DataFrame) – DataFrame containing cell properties.
image (numpy.ndarray) – Input image.
fluo_channel (int) – Index of the fluorescence channel.
label_matrix (numpy.ndarray) – Label matrix of the Voronoi diagram.
percentile (int) – Percentile to calculate for fluorescence intensity.

Returns:

DataFrame containing fluorescence properties for each cell.

Return type:

pandas.DataFrame

griottes.analyse.cell_property_extraction.voronoi_fluo_property_analysis(properties, image, mask_channel, radius, return_voronoi_mask=False, percentile=50)

Estimate the fluorescence properties of Voronoi cells.

Calculate the Voronoi mask for each cell, then use the mask to estimate the fluorescence intensities inside the mask.

Parameters:

properties (pandas.DataFrame) – DataFrame with nuclear properties, including centroid coordinates.
image (numpy.ndarray) – 3D image stack with fluorescence channels.
mask_channel (int) – Index of the nuclear mask channel.
radius (int) – Radius of the spherical mask used for Voronoi quantification.
return_voronoi_mask (bool, optional) – Whether to return the labeled Voronoi tesselation as a binary mask. Default is False.
percentile (int, optional) – Percentile used for fluorescence intensity quantification. Default is 50.

Returns:

DataFrame with fluorescence properties for each Voronoi cell, and optionally the labeled Voronoi tesselation as a binary mask.

Return type:

pandas.DataFrame or tuple of (pandas.DataFrame, numpy.ndarray)

Raises:

ValueError – If the Voronoi diagram cannot be computed.

griottes.analyse.cell_property_extraction.get_cell_properties(image, mask_channel=0, analyze_fluo_channels=False, fluo_channel_analysis_method='basic', cell_geometry_properties=False, labeled_voronoi_tesselation=False, radius=5, min_area=50, percentile=95, ndim=3)

Calculate the cell properties for a given image.

Parameters:

image (numpy array) – The image to be analyzed.
mask_channel (int) – The channel to be used as a mask.
analyze_fluo_channels (bool) – If True, the fluorescence channels will be analyzed.
fluo_channel_analysis_method (str) – The method to be used to analyze the fluorescence channels. Either basic, local_voronoi, or local_sphere.
cell_geometry_properties (bool) – If True, the cell geometry properties will be calculated.
labeled_voronoi_tesselation (bool) – If True, the voronoi tesselation will be generated and returned as an array.
radius (int) – Maximum radius within which the cell properties are measured.
min_area (int) – Minimum area of a cell to be considered.
percentile (int) – Percentile of the intensity distribution to be used for the percentile intensity calculation.

Return type:

pandas.DataFrame

voronoi_cell_property_analysis

griottes.analyse.voronoi_cell_property_analysis.in_hull(p, hull)

Test if points in p are in hull

p should be a NxK coordinates of N points in K dimensions hull is either a scipy.spatial.Delaunay object or the MxK array of the coordinates of M points in `K`dimensions for which Delaunay triangulation will be computed