A priori module definition

Home Forums General Help A priori module definition

This topic contains 2 replies, has 2 voices, and was last updated by  Thomas HINAULT 4 months, 2 weeks ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #25826 Reply

    Thomas

    Hi everyone,
    I would like to define modules a priori to focus on a specific set of regions. It is possible to manually check/uncheck nodes in the “edit community structure” window. However, this creates a new module each time a box is unchecked, resulting in a large number of modules..
    Is there a way to manually add/delete modules? And to keep one module fixed while running the Louvain algorithm on the rest of the network?
    Thanks,
    Thomas

    #26328 Reply

    Mite Mijalkov
    Keymaster

    Hi Thomas,

    Thank you for using Braph and I hope you find it useful in your research.

    The current implementation of modularity structure calculation forces every node to be a part of a module. That is why when you uncheck a node from a given module, it is automatically placed in a separate module at the end. However, when that node is checked in another, already existing module, the extra module is erased by the program, decreasing the number of modules.

    One way to avoid this problem is to first create the new module by unchecking one node. Then, the nodes that need to be part of this module can be directly added by checking the corresponding tick box (there is no need to uncheck them first, they will be immediately transferred). This should solve the problem of creating the new modules when unchecking the nodes.

    In order to keep one module fixed and then run community structure on the other ones, you would need to define a subgraph (consisting from the network minus the module you would like to keep fixed) an run your analysis on that graph.

    Hope this helps. If there is anything I can do more to help, please let me know.

    Best,
    Mite

    #26640 Reply

    Thomas HINAULT

    Ok this works.
    Thanks

    Thomas

Viewing 3 posts - 1 through 3 (of 3 total)
Reply To: A priori module definition
Your information: