Fig. 4: Distribution of brain fingerprint across resting-state functional networks.

Distribution of the edges with the highest ICC common to both cohorts (from ICC overlap binary matrix, cf. Fig. 3C) in within-networks and between-networks. A Distance from healthy reference computed as ratio disease/health, \({{\rm{R}}}\left({{\rm{net}}}\right)\), cf. “Methods”). Positive/negative values denote increase/decrease in the percentages of edges, respectively. Note, across all networks, the overall increase in percentages of edges in AD dementia patients and a slight decrease in MCI Aβ+. B Comparison of edges within- vs between-networks expressed as Chi-Square statistics. A high value with * denotes a significant (Bonferroni corrected) difference in the number of edges in within-networks vs between-networks. With some exceptions (see main text and Supplementary Fig. 2), this reflects a higher percentage of edges within relative to between networks. The chi-square for VIS in CU Aβ− and MCI Aβ+ was >800, but here we display a chi-square ≤ 300 for visualisation purposes. W within-networks, B between-networks, ICC intra-class correlation, CU Aβ− cognitively unimpaired Aβ-negative, VIS visual network, SMT somatomotor network, DA dorsal-attention network, SA salience network, L limbic network, FPN fronto-parietal network, DMN default-mode network, SBC subcortical regions.