Second Project for Network Science Course

This repository contains the code mentioned in the report which can be found here.

Dependencies

• bctpy: $pip install bctpy
• cdlib: $pip install cdlib
• scipy: $pip install scipy
• pandas: $pip install pandas
• matplotlib: $pip install matplotlib

Tested in python 3.7

Organization

/BrainConnectivityToolboxWrapper/

• Contains code to compute multiple metrics.
• More information on its own README file.

/data/pre-processed/

• The original data

/data/processed/

• Data after some transformations for our usage (file process.py). Includes edge list and matrices.
• Also includes some templates:
  • lobes.node: division of each lobe by color.
  • communities_template.node: coordinates of each of the 68 nodes.
  • template.node: template based on the Desikan atlas.

/DesikanViz/

• Populates the Desikan template for visualization.

/reports/

• Includes code to generate bar charts, box plots and scatter plots (plots.py).

Remaining folders are for saving the analysis results and statistics.

How to use

Run: $python main.py from root directory.

Parameters of function analyse may be changed, to select which metrics to run:

    degree_centrality = True/False
    node_strength = True/False,
    clustering_coeff = True/False,
    weighted_clustering_coeff = True/False,
    average_path_len = True/False,
    betweenness_centrality = True/False,
    newman_modularity = True/False,
    edge_betweenness = True/False,
    rich_club = True/False,
    participation_coefficient = True/False,
    communities_algorithm=
        'all'/'louvain'/'spinglass'/'walktrap'/'girvan_newman'/'infomap'
    desikan_metric=
        set this parameter with a node metric. (stats file header column)

Authors

• João Coelho
• João Saraiva