Bioenergetic Wiring Cost, Topological Hub Vulnerability, Developmental Lineage Origins, and In-Silico Therapeutic Rescue Across Three Adult Drosophila Connectomes (331,352 Neurons)

Correction notice (30 October 2026). A post-publication audit by the author found that several headline claims of this record are not supported. (1) The reported association between 3D wiring cost and "structural vulnerability" (ρ ≈ 0.85) is driven by synapse count, which enters both measures; after controlling for synapse count it vanishes (partial ρ ≈ −0.01). "Structural vulnerability" measures network centrality, not disease vulnerability. (2) In Paper II (10.5281/zenodo.23809545), the "Molecular Proteostasis Gap" contains wiring cost in its own formula, so its correlation with wiring cost is circular; the transcriptomic component alone is not significant (ρ = 0.17, p = 0.13). The channel-density correlation (ρ = 0.40) was with soma-to-arbor distance, not wiring cost (ρ = −0.11). (3) All in-silico "gene knockdowns", "therapies" and "rescues" in the LIF simulations (including the 97.7% mitophagy rescue, GABA potentiation and DBS) are direct consequences of hand-set model parameters (fraction of neurons removed, synaptic scaling factors) and contain no gene-specific biology; they must not be read as therapeutic predictions. (4) A pre-registered follow-up in human data (SEA-AD, 7 cortical regions) found that the fly-derived vulnerability indices do not predict neuronal loss in Alzheimer's disease. Descriptive connectome results (cross-connectome and cross-sex conservation, rich-club analysis, lineage descriptions) and the association between synapse count and Na+/K+-pump transcript allocation (partial ρ ≈ 0.51) are unaffected. The attached PDF still contains the original, unsupported claims. A corrected version is in preparation. Audit code and results are available from the author. Corrected abstract. We describe 3D synaptic wiring cost, network centrality and developmental lineage for 331,352 proofread neurons in three adult Drosophila melanogaster connectomes: FlyWire FAFB v783 (139,255 neurons), MaleCNS v1.0 (166,700) and the Janelia Hemibrain (25,397). Total synapse-weighted wiring cost of the female brain is 5,811 synapse·m. Per-neuron wiring cost and centrality (PageRank, degree) are both dominated by synapse count (ρ = 0.95 and 0.94), and are not related once synapse count is controlled for. Degree-preserving null models indicate an assortative core at intermediate degree and disassortative high-degree hubs. Synaptic load is conserved across 2,990 cell types shared by the three connectomes (ρ = 0.94), with 91% neurotransmitter concordance. Centrality differs between neuroblast hemilineages and between embryonic and post-embryonic neurons. In a leaky integrate-and-fire model, removing high-centrality neurons disrupts simulated sensorimotor activity more than random removal, a known property of hub-dominated networks; we present these simulations as illustrations of network sensitivity, not as models of disease or therapy. Whether these structural properties relate to disease vulnerability remains untested; a pre-registered test of related indices in human cortex was negative.