Metabolism-weighted brain connectome reveals synaptic integration and vulnerability to neurodegeneration

The remarkable abilities of the human brain arise from its specialized regions, which process and integrate information through complex connectivity patterns. While network science has developed various metrics to assess the degree of connectivity between these regions, it has not yet considered the level of activity within each region. Consequently, a highly connected region might be classified as a hub in the brain's connectome, even if it is weakly active or linked primarily to less active areas. Conversely, a region that is highly active but has only a few connections may be undervalued. To address this issue, we present a fully-weighted brain graph in which both edges (representing connectivity) and nodes (indicating metabolic activity) contribute to the importance, or centrality, of each region. In this model, the metabolism-weighted centrality (MwC) of each brain region integrates both connectivity and metabolic activity, using three datasets from simultaneously acquired functional MRI and metabolic FDG-PET data. We found that our fully-weighted brain graph demonstrates a greater ability to explain quantitative imaging data of signaling metabolism and its association with cognitive domains than a classical, edge-weighted graph. Additionally, regions with relatively high MwC exhibited increased synaptic and metabolic activity, as indicated by transcriptomic data. Furthermore, these regions showed higher susceptibility to neurodegenerative disorders. This framework of a fully-weighted brain graph represents a paradigm shift from connectivity-based metrics to an activity-aware brain graph. It offers a more biologically informed representation of brain network dynamics, connecting metabolic activity levels to higher cognitive functions as well as neurodegeneration.