conpagnon.plotting package¶
Submodules¶
conpagnon.plotting.display module¶
Display functions
Author: Dhaif BEKHA
-
conpagnon.plotting.display.display_gaussian_connectivity_fit(vectorized_connectivity, estimate_mean, estimate_std, raw_data_colors='blue', line_width=2, alpha=0.5, normed=True, bins='auto', fitted_distribution_color='black', title=None, xtitle=None, ytitle=None, legend_fitted='Fitted Gaussian Distribution', legend_data=None, display_fit='yes', ms=5)[source]¶ Display a vectorized connectivity matrices histogram, along with a Gaussian fit with an estimated mean, and standard deviation.
- Parameters
vectorized_connectivity (numpy.array, shape 0.5*n_features*(n_features + 1)) – A array of connectivity coefficient.
estimate_mean (float) – The estimated mean of the connectivity coefficient distribution.
estimate_std (float) – The estimated standard deviation of the connectivity coefficient distribution.
raw_data_colors (string, optional) – The colors of the histogram for the connectivity coefficient distribution. Default is blue.
line_width (int, optional) – The width of line for the fit of the data. Default is 2.
alpha (float, optional) – The opacity coefficient for histogram, between 0 and 1. Default is 0.5.
normed (bool, optional) – Normalized the histogram. This is mandatory for displaying the gaussian fit over the data, since it represent a probability density function.
bins (string, optional) – How the bins edges of the histogram are computed, choices among different estimators {‘fd’, ‘doane’, ‘scott’, ‘rice’, ‘sturges’} Default is ‘auto’, the maximum of the sturges and fd estimators are taken.
fitted_distribution_color (string, optional) – The color of the curves representing the Gaussian fit. Default is black.
title (string, optional) – The overall title of the figure. Default is None.
xtitle (string, optional) – The legend of the x-axis. Default is None.
ytitle (string, optional) – The legend of the y-axis. Default is None.
legend_fitted (string, optional) – The legend for the Gaussian fit. Default is ‘Fitted Gaussian Distribution’.
legend_data (string, optional) – The legend for the histogram of the data. Default is None.
display_fit (bool, optional) – If True, the gaussian fit of the data is displayed over the normalized histogram.
ms (float, optional) – The size of the dot on the graph, default is 5.
See also
matplotlib.pyplot.hist()This function, used here, compute and display the histogram of the data.
-
conpagnon.plotting.display.plot_matrix(matrix, labels_colors='auto', mpart='lower', k=0, colormap='RdBu_r', colorbar_params={'shrink': 0.5}, center=0, vmin=None, vmax=None, labels_size=8, horizontal_labels='auto', vertical_labels='auto', linewidths=0.5, linecolor='white', title='Untitled', figure_size=12, 9)[source]¶ Plot a entire matrix, or it’s lower part with a chosen heatmap.
- Parameters
matrix (numpy.array shape(numbers of regions, numbers of regions)) – The numerical array you want to plot.
labels_colors (str, or numpy.ndarray shape(number of regions, 3)) – The color in the RGB normalized space, that is each color are represented by triplet of float ranging between 0 and 1. Default is ‘auto’, random colors are automatically generated.
mpart (str) – The part of the array you want to plot, that is the entire matrix or just it’s lower part.
k (int) – The position from the main diagonal you want to cut. Default is 0.
colormap (str) – The seaborn colormap you want to apply on the matrix when displaying it. Default is ‘RdBu_r’.
colorbar_params (dict, optional) – Additional parameters concerning the parameters of the colorbar : position, size… Keyword arguments for fig.colorbar.
vmax (vmin,) – The extrema of the colormap. all display values are threshold according too those values. If None, these values are inferred from the matrix. Default is ‘None’.
labels_size (int, optional) – The font size of the labels. Default is 8.
horizontal_labels (str or list; optional) – The labels list of each rows. If ‘auto’, labels are replaced by the row index.
vertical_labels (str or list, optional) – The labels list of each columns. If ‘auto’, labels are replaced by the column index.
linewidths (float, optional) – The width that divide each cell. Default is 0.5.
linecolor (str, optional) – The color of the lines that divide each cell. Default is ‘white’.
title (str, optional) – The figure title. Default is ‘untitled’.
figure_size (tuple, optional) – Figure size in a tuple format (height, width). Default is (12, 9).
center (float, optional) – The center of the colormap, for divergent data.
See also
sns.heatmap()See the docstring of this function for a deeper insight of plotting function of a numerical array using seaborn.
-
conpagnon.plotting.display.plot_ttest_results(t_test_dictionnary, groupes, contrast, node_coords, node_color, output_pdf, node_size=50, colormap='bwr', colorbar=False, annotate=True, display_mode='ortho')[source]¶ Plot on a glass brain the significant mean effect resulting of a two samples t-test and save resulting figures in a PDF files.
- Parameters
t_test_dictionnary (dict) –
- A dictionnary containing multiple keys :
’tstatistic’ : The raw statistic t-map for the choosen contrast
2D numpy array of shape (number of regions, number of regions).
’uncorrected pvalues’ : The raw pvalues, 2D numpy array of shape
(number of regions, number of regions).
’corrected pvalues’ : The corrected pvalues with the chosen methods.
2D numpy array of shape (number of regions, number of regions).
’significant edges’ : The significant t-values after masking
for the non significant pvalues at alpha level. 2D numpy array of shape (number of regions, number of regions).
’significant pvalues’ : The significant pvalues at level alpha.
2D numpy array of shape (number of regions, number of regions)
’significant mean effect’ : The differences of mean connectivity
between the two groups according to the chosen contrast, and non-significant connexion are mask at alpha level. 2D numpy array of shape (number of regions, number of regions).
groupes (list) – The list of groupes involved in the computed contrast.
contrast (list) – The contrast vector used in the two samples t-test.
colormap (str, optional) – The colormap to use for the matrix heatmap. Default is ‘bwr’.
node_coords (list) – The list of centroids coordinates of each regions in the atlas used for regions signals extraction.
node_color (numpy.ndarray shape(number of regions, 3)) – The color in the RGB normalized space, that is each color are represented by triplet of float ranging between 0 and 1.
output_pdf (str) – The full path including .pdf extension to the a pdf file for saving the glass brain projection plot.
node_size (int, optional) – The node size. Default is 50.
colorbar (bool, optional) – If True, the colorbar of the chosen colormap is displayed alongside the glass brain plot. Default is False.
annotate (bool, optional) – If True, letters at the top of the glass brain indicates the left or right hemisphere location. Default is True.
display_mode (str, optional) – The slice configuration you want to display Possible values are: ‘ortho’,’x’, ‘y’, ‘z’, ‘xz’ ‘yx’, ‘yz’, ‘l’, ‘r’, ‘lr’, ‘lzr’, ‘lyr’,’lzry’, ‘lyrz’. Default if ‘ortho’.
See also
parametric_tests.two_sample_t_test()This function return a dictionnary resulting of the comparison of two groupe using a t-test. This the input of this display function.
nilearn.plotting.plot_connectome()This function in the Nilearn packages is used here for the plot of the glass brain. I encourage the user to read the corresponding docstring in the Nilearn git repo.
Notes
The variables groupes and contrast are needed only the title. You shoul have already used this variables when you compute the two sample t-test between you’re two groups.
-
conpagnon.plotting.display.t_and_p_values_barplot(t_values, p_values, alpha_level, xlabel_color, bar_labels, t_xlabel, t_ylabel, p_xlabel, p_ylabel, t_title, p_title, xlabel_size=10, ylabel_size=10, size_label=8)[source]¶ Plot a barplot representation of t and p values resulting of a two sample Student t-test in two separates figure.