Data and code from: Spatial and morphological organization of mitochondria in neurons across a connectome

Neuronal function depends on mitochondria, but little is known about their organization across neurons. Using an electron microscopy Drosophila connectome, we uncovered quantitative rules governing the morphology and positioning of hundreds of thousands of mitochondria across thousands of neurons. We discover that mitochondrial morphological features are specific to cell and neurotransmitter type, providing fingerprints to identify neurons. Mitochondria are positioned with 2–3 µm precision relative to synaptic and structural features, with systematic differences across neuron types and compartments. Mitochondrial localization correlates with regional activity and postsynaptic targets. Analysis of a mouse visual cortex connectome confirms cell-type specific morphology and identifies partially divergent positioning rules. These results establish mitochondria as circuit-embedded organelles whose distribution links subcellular architecture to brain connectivity.

Methods
Software and datasets

Code to recreate these analyses is available at: https://github.com/ClarkLabCode/MitochondrialMorphologyPosition. This code is built upon the open source neuprint software (91) to access the hemibrain dataset (1), in particular the segmented mitochondria (26).