interfaces.cmtk.nbs

NetworkBasedStatistic

Link to code

Calculates and outputs the average network given a set of input NetworkX gpickle files

For documentation of Network-based statistic parameters:

Example

>>> import nipype.interfaces.cmtk as cmtk
>>> nbs = cmtk.NetworkBasedStatistic()
>>> nbs.inputs.in_group1 = ['subj1.pck', 'subj2.pck'] 
>>> nbs.inputs.in_group2 = ['pat1.pck', 'pat2.pck'] 
>>> nbs.run()

Inputs:

[Mandatory]
in_group1: (a list of items which are an existing file name)
        Networks for the first group of subjects
in_group2: (a list of items which are an existing file name)
        Networks for the second group of subjects

[Optional]
edge_key: (a unicode string, nipype default value: number_of_fibers)
        Usually "number_of_fibers, "fiber_length_mean", "fiber_length_std"
        for matrices made with CMTKSometimes "weight" or "value" for
        functional networks.
ignore_exception: (a boolean, nipype default value: False)
        Print an error message instead of throwing an exception in case the
        interface fails to run
node_position_network: (a file name)
        An optional network used to position the nodes for the output
        networks
number_of_permutations: (an integer (int or long), nipype default
         value: 1000)
        Number of permutations to perform
out_nbs_network: (a file name)
        Output network with edges identified by the NBS
out_nbs_pval_network: (a file name)
        Output network with p-values to weight the edges identified by the
        NBS
t_tail: (u'left' or u'right' or u'both', nipype default value: left)
        Can be one of "left", "right", or "both"
threshold: (a float, nipype default value: 3)
        T-statistic threshold

Outputs:

nbs_network: (an existing file name)
        Output network with edges identified by the NBS
nbs_pval_network: (an existing file name)
        Output network with p-values to weight the edges identified by the
        NBS
network_files: (a list of items which are an existing file name)
        Output network with edges identified by the NBS

ntwks_to_matrices()

Link to code