The Connectome Analyzer also provides extra features for P-value matrix relaxation and compute Single subject measures
Matrix file
Select the file containing the matrix of p-values to relax. The file can be in csv format or matlab ”.mat” format. In the case of a ”.mat” file, make sure that the variable containing the p-values is named “matrix”. The p-values can come from any test (Student t-test, Wilcoxon test, multivariate Hotelling test, etc...), and should be ordered as a single column in the file
Data separator
If the p-value file is in csv format, select the character that separates columns in the file. For matlab ”.mat” files, this setting is ignored
Decomposition method
- Automatic: This feature should actually not be used for single p-values relaxation. Set the decomposition manually using the file option.
- File: provide a file containing the decomposition of the p-values. The file should have 2 columns and as much rows as p-values. The first column should contain the p-value label, ranging from 1 to the maximum number of p-values. The second column should contain the sub-region to which the corresponding p-value belongs. Sub-regions labels can range from 1 to the number of sub-regions, without skipping numbers
Correction
Select the method used for the correction for multiple testing
Alpha1
Threshold to detect significantly different sub-networks
Alpha2
Threshold bellow which the null hypothesis Group1 == Group2 can be rejected in favor of the H1 used to compute the original p-values
Strong
Factor by which the p-values of non-significatively different sub-networks should be multiplied in the RMIO procedure
Connectivity matrix
Select the file containing the connectivity matrix. The file can be in csv format or matlab ”.mat” format. In the case of a ”.mat” file, make sure that the matrix variable is named “matrix”.
Data separator
If the connectivity matrix file is in csv format, select the character that separates columns in the file. For matlab ”.mat” files, this setting is ignored
Network measures
- You can choose to compute the network measures on:
- Non-Weighted graphs: the edge values represent the number of connections between the vertices
- Weighted graphs: the edge values represent some measure of the connection strength between the vertices
- Binarized graphs: the edges that have non-zero values are set to one before computing the network measures
- Select the set of measures you want to test:
- Degree: computes the mean degree of the network. If the matrix is weighted, the mean strength is computed
- Betweenness: computes the network mean node betweenness
- Closeness: computes the network mean closeness
- Diameter: computes the network mean diameter
- Efficiency: compute the network mean efficiency