Diagnostics of network topology in real and synthetic network models.

In the second column, we report mode degree and the number of nodes identified with that degree (note that the mean degree is identical for all synthetic and real networks by construction). In the remaining columns, we report the mean and standard deviation of network diagnostics values, calculated over 100 realizations of each synthetic network model. We report the empirical value for the brain data (see Supplementary Material for additional results for other participants). Note that the algorithms for determining fractal dimension and Rentian scaling are not deterministic and therefore multiple estimates can be obtained by applying the algorithms to the same network multiple times. To assess the relative role of different types of variance in the estimated value of diagnostics, we provide Table S2 in the Supplementary Materials, where we report confidence intervals in the fitted values of several diagnostics (fitting variance), standard deviation of the diagnostic value over 100 computations (algorithmic variance), and standard deviation of the diagnostic value over 100 realizations of the network (model variance).