Brain-wide Organization of Post-Synaptic Sites: Three Principles

Abstract Synapses are fundamental units of neural computation and structural plasticity in the brain, yet their brain-wide spatial organization in mammals remains largely unknown. Building on our previous mapping and modeling of dendritic spines and axonal boutons, we developed a statistical approach to generate over 17.99 million predicted post-synaptic sites based on the potential arbor contacts (PPSSPAC) across the dendritic arbors of 155,743 neurons spanning the entire mouse brain. By analyzing the topological features of these sites, we identified three key organizational principles. First, PPSSPAC exhibits a distinct region-specific spatial pattern at the dendritic branch level, preferentially enriching near the distal ends of dendritic segments, in contrast to the valley-like distribution of pre-synaptic sites along axonal branches. Cross-species validation against independent electron microscopy datasets from mouse and human brains, revealed a conserved and previously unquantified pattern of synaptic organization. Second, the spatial distribution of PPSSPAC effectively captures variations across cell types in the hippocampus, enabling neuron classification based on synaptic architecture. This serves as a complementary mesoscale descriptor to existing classifications based on global neuronal connectivity (source-target relationships) or local dendritic microenvironments. Third, the spatial distribution of PPSSPAC, rather than synapse density, shows remarkable correlation with independent whole-brain electrophysiological activity. The spatial organization of PPSSPAC also correlates with several gene expression datasets. These results suggest the post-synaptic pattern could be mediator linking molecular signatures, neural dynamics, and connectivity types of cells. Our study reveals, for the first time, a structural substrate for hierarchical information processing in the brain, providing new insights into neural computation and a foundation for brain digital twin models to support next-generation AI infrastructure. Competing Interest Statement The authors have declared no competing interest.