Sign in

Abstract

Evolution has shaped nervous systems to produce stereotyped behavioural responses to ethologically relevant stimuli. For example when laying eggs, female Drosophila avoid geosmin, an odorant produced by toxic moulds. Here we identify second, third, and fourth order neurons required for this innate olfactory aversion. Connectomics data place these neurons in a complete synaptic circuit from sensory input to descending output. We find multiple levels of valence-specific convergence, including a novel form of axo-axonic input onto second order neurons conveying another danger signal, the pheromone of parasitoid wasps. However, we also observe extensive divergence: second order geosmin neurons connect with a diverse array of 80 third order cell types. We find a pattern of convergence of aversive odour channels at this level. Crossing one more synaptic layer, we identified descending neurons critical for egg-laying aversion. Our data suggest a transition from a labelled line organisation in the periphery to a highly distributed central brain representation that is then coupled to distinct descending pathways.

PDFDOI
Structured summary ↓

02 /Specification & Classification

Click any tag to browse matching papers

Abstract summary

Verified abstract synthesis

Claude Opus 5.5 · xhigh effort · 18 Oct 2026 · from the abstract · CC BY-NC 4.0

Geosmin avoidance in Drosophila runs from a labelled line at the periphery into a distributed central representation spanning 80 third-order cell types, then into specific descending neurons.

Nervous systems produce stereotyped behavioural responses to ecologically relevant threats, such as egg-laying female Drosophila avoiding the toxic-mould odorant geosmin. Tracing this pathway across every synaptic layer from sensory input to descending output reveals how a peripheral labelled line transitions into a distributed central brain representation where danger signals converge.

Key results

  1. SuggestedA continuous circuit for one innate aversion is assembled, from the sensory channel through second-, third- and fourth-order neurons to descending neurons tied to egg-laying avoidance.

    Neurons said to be required for the aversion were identified at each order and positioned in a complete synaptic chain using connectomics data; the abstract names neither the connectome resource nor the functional assays. · Abstract

  2. ShownThe channel does not stay narrow: it fans out onto a large and heterogeneous set of central cell types one synapse in.

    Second order geosmin neurons connect with 80 third order cell types, which the authors read as a move from peripheral labelled lines to a distributed central brain representation. · Abstract

  3. SuggestedSeparate danger signals also meet, with aversive odour channels converging and novel axo-axonic input onto second-order neurons conveying another danger signal, the pheromone of parasitoid wasps.

    Reported as valence-specific convergence at several levels, with the axo-axonic arrangement described by the authors as a new form; no synapse counts or statistics are given in the abstract. · Abstract

Methods, data and reproducibility

Neurons needed for innate avoidance of geosmin were identified at the second, third and fourth order of the Drosophila olfactory pathway, and connectomics data were used to place them in one synaptic circuit reaching from sensory input to the descending neurons linked to egg-laying aversion. The abstract names no connectome dataset, imaging protocol or behavioural assay.

  • Connectomics data used to reconstruct the synaptic circuit between sensory input and descending output · Drosophila
Limitations (2)
  • Neither the connectome volume nor the functional and behavioural assays nor the number of animals is stated, so the evidence behind calling neurons required or critical cannot be assessed from this text. · noted in this summary
  • Convergence and divergence are summarised as a cell-type count and qualitative patterns; no synapse numbers, statistics or behavioural effect sizes are reported. · noted in this summary

Who should read:Read this if you study olfactory circuits or connectome-guided tracing and want a worked case of a labelled line dissolving into a distributed central representation.

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.