5 to 100% using increments of 5 (Fig. 5). For lower thresholds, the network tends to split in two or more subnetworks, resulting in the lower C and shorter L. As the threshold increases, both C and L also rapidly increase until the network is fully connected. In this study, the full connection was predicted based on the Erdös–Rényi model (Erdös and Rényi, 1961) of random graphs, where a graph with N nodes is fully connected if the connection density is larger than 2lnN/N (≈ 23.2% in this study). For thresholds higher than this value, the network increasingly includes weak and spurious connections, and the value of network metrics tends to saturate or slightly decrease. We found consistent network alterations for global network measures using the varying threshold, while the threshold was fixed with 30% of connection density for local measures and correlations since the networks have the maximum C and L at that value (Fig. 5).
