Dated snapshot of an independent research project on the fruit-fly connectome. Whole-brain models built on the FlyWire connectome need a sign (excitatory or inhibitory) for every neuron; these signs come from a machine classifier and a convention. In a published whole-brain model (Shiu et al. 2024), one olfactory receptor neuron activates 97% of the antennal-lobe projection neurons (first observed 24 September 2026). The cause is the default given to the 150 antennal-lobe local neurons that receive no transmitter prediction ("no call → excitatory"). The model's sense of smell depends on how much of the local neurons' output is excitatory, not on which cells carry it. Given a real fly's receptor responses (Seki et al. 2017), the published signs reproduce the real projection-neuron responses at r = 0.28, a laboratory-informed prior at r = 0.70. Every test was pre-registered in a hash-frozen file, and every number is linked to the script output that prints it. Work in progress, not peer reviewed.
No takes yet. Share an insight, caveat, or question.
Shakib hawoladher (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: