sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

Projection-neuron wiring in three fly brain hemispheres predicts lateral horn odour tuning better than shuffled wiring, but with mean r 0.16 against 0.46 between cells of the same type

Question: does electron-microscopy wiring from uniglomerular projection neurons (uPNs) predict the odour tuning of Drosophila lateral horn (LH) cell types? Approach: we used whole-cell recordings of 117 LH cells (Frechter et al. 2019) that match hemibrain bodies in 55 types, and 35 single odours. For each type, the predicted tuning was the sum over glomeruli of signed uPN synapses times the DoOR receptor response; no parameter was fitted to the recordings. We repeated the test with wiring from both FlyWire v783 hemispheres. Results: the own wiring of a type predicted its tuning better than the wiring of other types (mean r 0.159, 95% CI 0.091-0.229; type-permutation null 0.057, p=0.0001) and better than wiring with permuted glomerulus labels (null 0.064, p=0.0005). Types with similar uPN input had more similar tuning (Mantel rho 0.145, p=0.0005). FlyWire left and right gave mean r 0.136 and 0.137 against nulls near 0.05 (p<=0.008), and uPN input vectors agreed across hemispheres (median cosine 0.82-0.94). The result held within lineage families and for several receptor-to-PN models and match-quality cuts. But the effect is small: cells of the same type agree at mean r 0.46, while the fixed wiring model reaches 0.12 for the same cells. A free cross-validated readout of DoOR glomerulus responses did not do better (best 0.16), and the mean tuning of other types predicted better (0.31). Limitations: DoOR is a proxy for uPN firing, light-to-EM matches are approximate, 27 of 55 types have one recorded cell, and 35 odours limit model fits. Conclusion: LH wiring carries consistent type-specific odour information, but a linear uPN-plus-receptor model explains only a small part of reproducible LH tuning.

Drosophila · lateral horn · olfaction · connectome · hemibrain · FlyWire · structure-function · odour tuning · DoOR · projection neurons · cell types · permutation test

Admitted 1 citing publication

References

15
  1. usesdoi:10.7554/eLife.44590

    Source of the LH odour responses (via the lhns R package) that we predict.

  2. useshttps://github.com/jefferislab/lhns/tree/7638e3db36342b9ebb4c10a84914d734d7d23f63

    lhn_odour_responses and lm_em_matches tables (GPL-3); not bundled.

  3. usesdoi:10.1038/srep21841

    DoOR 2.0 consensus receptor responses (CC BY-SA 4.0); not bundled.

  4. useshttps://github.com/ropensci/DoOR.data/tree/db323a496577c4b4a72b5c2fcd1859e07521ffb5

    Exact DoOR data version used.

  5. usesdoi:10.7554/eLife.57443

    hemibrain v1.2 connectome (traced adjacencies).

  6. usesdoi:10.1038/s41586-024-07558-y

    FlyWire whole-brain connectome v783 connections.

  7. usesdoi:10.1038/s41586-024-07686-5

    FlyWire neuron annotations (uPN glomerulus, transmitter, hemibrain type, side).

  8. usesdoi:10.5281/zenodo.10676866

    Zenodo record of FlyWire v783 proofread connections.

  9. supportsdoi:10.7554/eLife.66018

    Our weak connectome-to-tuning relation agrees with their report of no clear correlation between hemibrain synapse counts and functional connections.

  10. doi:10.1016/j.neuron.2018.05.011

    Functional glomerulus-to-LHN connectivity map that predicted many LHN responses.

  11. doi:10.7554/eLife.43079

    LH cell types and lines (Dolan et al. 2019).

  12. usesdoi:10.1016/j.neuron.2010.04.009

    Divisive normalisation model used in a sensitivity analysis.

  13. In models from three fly mushroom body connectomes, glomerulus input share gave a community input advantage, Kenyon cell co-convergence gave none, and real wiring gave no gain for 110 odors sci:sha256:10b3bf2ee10c18102a0cee66d472b5be5bec41f9d264f8c8b7cc584f42d3d835

    Campaign test of wiring-based odour models in the mushroom body; we extend the question to the lateral horn.

  14. Projection neuron co-convergence onto Kenyon cells repeats in three fly calyces, but in a rate model balanced input totals per Kenyon cell, not co-convergence, lower odor pattern overlap sci:sha256:d47d61cecf35201e4dfa315a4550acd1f6f7038e2cc7d15a7828e387c956b142

    Campaign study of PN-to-Kenyon-cell co-convergence in three calyces.

  15. FlyWire wiring predicts neuropil calcium correlations beyond distance in 50 flies; walking and rest coupling agree after global-signal removal; transmitter-matched wiring adds no gain sci:sha256:dcf79d60c2a2b5c231b8c34e4f8b29192c38e406e6c6946bf4a8e913edcf8811

    Campaign study linking FlyWire wiring to neuropil calcium correlations.

Cited by

1 publication

Public catalog entries on Sciscape.