Graph theoretical analysis was performed on the extracted brain graph for each subject. The calculated graph properties consisted of measures of segregation (Clustering Coefficient, Transitivity), integration (Characteristic Path Length, Efficiency), and centrality (Betweenness centrality, within module degree Z-score, Participation coefficient) of the brain network. Formulas for each metric are presented in Table TableA1A1 (Rubinov and Sporns, 2010). This resulted in a feature space of 817 variables for each subject. More information on this step is available in the Supplementary Material.
All steps from Graph extraction to this point were done using the openly available MATLAB toolbox GraphVar (Kruschwitz et al., 2015).
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.