neurostatx.cli.VisualizeGraphNetwork

Attributes

Name	Description
app	Cyclopts application for the VisualizeGraphNetwork command-line tool.

Functions

Name	Description
VisualizeGraphNetwork	Visualize a graph network generated from clustering results.

appattributemodule attribute#

View source

app = App(default_parameter=Parameter(negative=()))

Cyclopts application for the VisualizeGraphNetwork command-line tool.

VisualizeGraphNetworkfunction#

View source

def VisualizeGraphNetwork(in_graph: Annotated[str, Parameter(show_default=False, group='Essential Files Options')], out_folder: Annotated[str, Parameter(show_default=True, group='Essential Files Options')] = './visualize_network/', verbose: Annotated[bool, Parameter(-v, --verbose, group='Optional parameters')] = False, overwrite: Annotated[bool, Parameter(-f, --overwrite, group='Optional parameters')] = False, save_parameters: Annotated[bool, Parameter(-s, --save_parameters, group='Optional parameters')] = False, label_name: Annotated[List[str], Parameter(show_default=False, group='Label Options')] = None, cohort: Annotated[int, Parameter(show_default=False, group='Essential Files Options')] = None, background_alpha: Annotated[bool, Parameter(show_default=True, group='Label Options')] = True, weight: Annotated[str, Parameter(show_default=False, group='Network Visualization Options')] = 'membership', label_centroids: Annotated[bool, Parameter(show_default=True, group='Network Visualization Options')] = True, label_subjects: Annotated[bool, Parameter(show_default=True, group='Network Visualization Options')] = False, centroids_size: Annotated[int, Parameter(show_default=True, group='Network Visualization Options')] = 500, centroid_alpha: Annotated[float, Parameter(show_default=True, group='Network Visualization Options')] = 1, centroid_node_color: Annotated[str, Parameter(show_default=True, group='Network Visualization Options')] = 'white', centroid_edge_color: Annotated[str, Parameter(show_default=True, group='Network Visualization Options')] = 'black', subject_node_size: Annotated[int, Parameter(show_default=True, group='Network Visualization Options')] = 5, subject_node_alpha: Annotated[float, Parameter(show_default=True, group='Network Visualization Options')] = 0.1, subject_node_color: Annotated[str, Parameter(show_default=True, group='Network Visualization Options')] = 'darkgrey', subject_edge_color: Annotated[str, Parameter(show_default=True, group='Network Visualization Options')] = None, colormap: Annotated[str, Parameter(show_default=True, group='Network Visualization Options')] = 'plasma', title: Annotated[str, Parameter(show_default=False, group='Network Visualization Options')] = 'Network Graph of the clustering membership values.', legend_title: Annotated[str, Parameter(show_default=False, group='Network Visualization Options')] = 'Membership values')

Visualize a graph network generated from clustering results.

VisualizeGraphNetwork uses NetworkX to draw the graph and can label specific nodes based on a condition (for example a diagnosis). Appearance can be customized with the visualization parameters. It is meant to work with ComputeGraphNetwork because it requires precomputed node positions in the graph file.

Notes

Labelling nodes

Specific nodes can be labelled from a binary node attribute via –label-name. Multiple labels can be provided with repeated –label-name flags; the script writes one graph per label name.

Cohort selection

When the graph contains data from multiple cohorts, –cohort fetches the specified cohort and filters –label-name. The graph must contain a cohort attribute; add it with AddNodesAttributes if needed.

Graph appearance

Using subjects labelling will crowd the network if it contains a high number of nodes. Centroids are labelled by default c1, c2, ... and subjects s1, s2, .... The script also exports a graph_network_file.gexf file that can be customized further in other tools such as Gephi [1].

References

[1] Gephi

Parameters

Name	Type	Default	Description
in_graph	str	required	Input graph network to visualize (.gml format).
out_folder	str	'./visualize_network/'	Output folder for the graph network visualization. Defaults to ./visualize_network/.
verbose	bool	False	Verbose mode. Defaults to False.
overwrite	bool	False	Overwrite existing files. Defaults to False.
save_parameters	bool	False	Save parameters to a .txt file. Defaults to False.
label_name	List[str]	None	List of label names to subsequently use for labelling. Defaults to None.
cohort	int	None	Cohort identifier. If your graph contains data from multiple cohorts, you can specify the cohort you want for visualization. Defaults to None.
background_alpha	bool	True	Use background alpha for the graph. Defaults to True.
weight	str	'membership'	Weight to use for the graph network. Defaults to membership.
label_centroids	bool	True	Label centroids. Defaults to True.
label_subjects	bool	False	Label subjects. Defaults to False.
centroids_size	int	500	Size of the centroids. Defaults to 500.
centroid_alpha	float	1	Alpha of the centroids. Defaults to 1.
centroid_node_color	str	'white'	Color of the centroids. Defaults to white.
centroid_edge_color	str	'black'	Edge color of the centroids. Defaults to black.
subject_node_size	int	5	Size of the subjects. Defaults to 5.
subject_node_alpha	float	0.1	Alpha of the subjects. Defaults to 0.1.
subject_node_color	str	'darkgrey'	Color of the subjects. Defaults to darkgrey.
subject_edge_color	str	None	Edge color of the subjects. Defaults to None.
colormap	str	'plasma'	Colormap to use for the graph network. Defaults to plasma.
title	str	'Network Graph of the clustering membership values.'	Title of the graph network. Defaults to Network Graph of the clustering membership values..
legend_title	str	'Membership values'	Legend title of the graph network. Defaults to Membership values.

Examples

VisualizeGraphNetwork --in-graph graph_network.gml
--out-folder output/ --label-name diagnosis
--weight membership -v -f -s