ConnectionMiner · type-pair matrices
Three independent 741×741 matrices over the same catalogue of named Drosophila visual-system cell types, indexed in the same alphabetical order so a row/column lines up across all panels. C is the binary synaptic connectome (does type i synapse onto type j, pooled across both hemispheres). d(soma) is the mean soma-to-soma Euclidean distance between every cell of type i and every cell of type j, measured from just the root/soma node of each cell's traced SWC skeleton. d(neurite) is the true closest-approach distance instead — the minimum distance between ANY point on ANY neurite of type i and ANY point on ANY neurite of type j, using the full skeleton (every traced point, no subsampling) rather than just the cell body, so it reflects how close the actual axon/dendrite arbors get to each other, not just how far apart the somas sit. Type-pairs whose bounding boxes don't overlap get an exact brute-force minimum; for pairs still pending full brute force (a GPU-time-limited, checkpointed, resumable computation — see cell hover), the value shown is instead an exact, mathematically guaranteed LOWER bound from the two types' bounding-box gap — never an approximation or random-sampled estimate, just not yet the precise minimum.
1 if the EM-reconstructed connectome contains ≥1 synapse from type i onto type j, else 0.
Mean Euclidean distance between soma positions of all cells of type i vs. all cells of type j (self-pairs excluded on the diagonal).
Minimum distance between any neurite point of type i cells and any neurite point of (different) type j cells — the closest their arbors ever get. Hover a cell to see whether that value is the exact minimum or an exact lower bound (still pending full computation).