Connectome-Constrained Spiking Neural Networks: Olfactory Classification Study

By Vicktor Moberg and Madelyn Pierce

Biological brains are wired by millions of years of evolution, while artificial neural networks learn over largely arbitrary topologies. Does that evolved wiring actually encode a computational advantage? To find out, we built a spiking neural network constrained to the real olfactory circuit of the fruit fly Drosophila melanogaster — mapped synapse-by-synapse by the FlyWire connectome — and tested it against randomly-wired and conventional network baselines on an odor-classification task. Our finding: under these conditions, biological topology performed no better than a degree-matched random control, narrowing where the advantages of evolved circuits are likely to lie.

This work reflects MIRE's commitment to rigorous, open, and honest science — including the publication of clear negative results, which are as valuable to the field as positive ones. By releasing our full code, connectome-extraction pipeline, and trained weights, we aim to support reproducibility and invite others to extend the work to new circuits and learning rules.

Read the full paper below and explore the code on GitHub — https://github.com/VickM12/flywire-olfactory-snn