Fig. 1: PLSC results: Saliences and Scores.
From: Brain metabolic-functional (de)coupling from health to glioma dysfunction
a S and SUVR PLSC saliences, over the bootstrap samples, only the PLSC salience weights significantly different from zero are reported. b \(\widehat{{\mbox{FC}}}\) and MC network-wise PLSC salience pattern. These matrices are obtained by averaging the entries values in each block, considering zero the non-significant entries. c Scatter dots plotted in the latent space for both S–SUVR (left) and \(\widehat{{\mbox{FC}}}\)–MC (right), where each point corresponds to a healthy control’s score, alongside the least-square best-fitting line. The colored boxes highlight the subjects who primarily contribute to defining the variance of the component, exhibiting distinct metabolic-functional characteristics. Vis, visual network; SomMot, somatosensory motor network; DorsAttn, dorsal attention network; SalVentAttn, saliency/ventral attentional network; Limbic, limbic network; Cont, control network; DMN, default mode network, Subcortical, subcortical regions: thalamus proper, caudate, putamen, pallidum, cerebellum, and hippocampus.
