Figure 5.

Activity dependent modification of the prevalence of synaptic cluster morphology and organization. (A,B) Hierarchical grouping of presynaptic Bassoon (A) and postsynaptic PSD95 (B) cluster subtypes for 4 selected features (hierarchical grouping including all features are shown in Supplementary Figures 16, 17). Synaptic subtypes were identified from the detected local maxima in the KDE plots generated from the scatterplot of the UMAP embedding of all detected Bassoon and PSD95 clusters (Figure 4, Supplementary Figure 13). Minimum (yellow) and maximum (dark blue) value of the heatmap for each feature: coupling probability (min: 0, max: 1), distance (min: 0 nm, max: 180 nm), area (min: 0.7·10⁻² μm², max: 11.8·10⁻² μm²), and eccentricity (min: 0.4, max: 1). (C,D) Representative STED images of (C) Bassoon and (D) PSD95 clusters (top) with corresponding two-color STED image (bottom). Proportions of Bassoon (E,G) and PSD95 (F,H) cluster belonging to each synaptic subtype depending on the neuronal activity state: high Mg²⁺/low Ca²⁺ vs. 0Mg²⁺/Gly/Bic (E,F), high Mg²⁺/low Ca²⁺ vs. Glu/Gly (G,H). Stars represent significant changes in the proportion of synaptic subtypes using Chi-square test. Exact p-values are reported in Supplementary Tables 2, 4 with *p < 0.05, **p < 0.01, and ***p < 0.001. Image size: 0.96 μm².