Standardized FlyWire skeletons reveal neuropil-specific organization of synaptic topology
Year: 2026
Type: article
Abstract: Reliable neuronal skeletons are essential for large-scale morphological and connectomic analyses, yet automated skeletonization can introduce local structural and topological artifacts. Here, we systematically assessed 139,273 FlyWire v783 skeletons and implemented a human-in-the-loop framework to refine soma-associated artifacts, standardize branching topology, and integrate synapse annotations. The refined skeletons preserved global morphology and neuronal identity, with a median normalized NBLAST similarity of 0.992 and 100% Top-1 identity retention. Using this resource, we identified neuropil-specific branch-order organization of synapse-associated nodes. Nodes associated exclusively with postsynaptic sites were located at higher mean branch orders than those associated exclusively with presynaptic sites in 39 of 43 neuropils. This pattern remained robust without multifurcation binarization. Our study provides a standardized, quality-controlled FlyWire skeleton resource for brain-wide analyses of neuronal morphology and synaptic topology.
Source: Brain Informatics
Authors: Longxiao Yuan, Zhiyan He, Hanchuan Peng
Institutions: Fudan University, Columbia University, Barnard College
Language: English
Subfield: Biophysics
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $1,090
OpenAlex ID: W7225043632