An information-theoretic analysis of the tsetse fly ( Glossina morsitans ) antenna reveals trichoid sensilla as the main carriers of odor identity

Tsetse flies locate vertebrate hosts largely by smell, and odor-baited devices are central to tsetse and trypanosomiasis control. A published electrophysiological survey of the Glossina morsitans antenna defined seven classes of olfactory sensilla, but descriptive clustering cannot reveal how much information these classes carry or which odors they distinguish. We reanalyzed responses of 182 sensilla to 16–17 odorants using Fisher–Rao information geometry, bias-corrected mutual information, Bayesian decoding, and signal detection theory. In the fruit fly Drosophila melanogaster , trichoid sensilla are largely dedicated to pheromones. In tsetse, the three trichoid classes carried more information about odor identity than the four basiconic classes (1.92 vs. 1.14 bits, of a possible 4), a difference of 0.78 bits (95% confidence interval 0.44–1.10), and were strongly excited by the host attractant 1-octen-3-ol. Adding trichoid to basiconic sensilla raised decoding accuracy from 32% to 70% (23% to 55% for decoders tested on held-out sensilla) and identified 1-octen-3-ol on 98% of trials (92% held out). The basiconic code was nearly one-dimensional (participation ratio 1.71), organized along an axis of response strength with 1-octen-3-ol at its extreme. The CO 2 -sensitive ab4 class was the most sensitive detector for 11 of 16 odorants, whereas ab2 carried no detectable odor information. Classes shared up to 0.41 bits of their information with one another, an overlap only weakly predicted by the similarity of their mean tuning. On a shared odor panel, Drosophila receptors carried twice the information per channel and, with channel number matched, spanned about 1.5-fold more dimensions. These conclusions held in simulations at the real sample sizes, under alternative noise models, and when classes were defined on independent odorants. The tsetse antenna thus combines redundant, detection-oriented basiconic channels with trichoid sensilla that carry host-odor identity, principles relevant to the design of odor baits.

Authors

Institutions

Publication Details

Journal
bioRxiv (Cold Spring Harbor Laboratory)
Published
2026-10-28
DOI
https://doi.org/10.64898/2026.10.26.761766
Primary Topic
Neurobiology and Insect Physiology Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

An information-theoretic analysis of the tsetse fly ( Glossina morsitans ) antenna reveals trichoid sensilla as the main carriers of odor identity

Neeraj Soni
bioRxiv (Cold Spring Harbor Laboratory)
Neurobiology and Insect Physiology Research
preprint

An information-theoretic analysis of the tsetse fly ( Glossina morsitans ) antenna reveals trichoid sensilla as the main carriers of odor identity

Neeraj Soni
preprint en

Abstract

Tsetse flies locate vertebrate hosts largely by smell, and odor-baited devices are central to tsetse and trypanosomiasis control. A published electrophysiological survey of the Glossina morsitans antenna defined seven classes of olfactory sensilla, but descriptive clustering cannot reveal how much information these classes carry or which odors they distinguish. We reanalyzed responses of 182 sensilla to 16–17 odorants using Fisher–Rao information geometry, bias-corrected mutual information, Bayesian decoding, and signal detection theory. In the fruit fly Drosophila melanogaster , trichoid sensilla are largely dedicated to pheromones. In tsetse, the three trichoid classes carried more information about odor identity than the four basiconic classes (1.92 vs. 1.14 bits, of a possible 4), a difference of 0.78 bits (95% confidence interval 0.44–1.10), and were strongly excited by the host attractant 1-octen-3-ol. Adding trichoid to basiconic sensilla raised decoding accuracy from 32% to 70% (23% to 55% for decoders tested on held-out sensilla) and identified 1-octen-3-ol on 98% of trials (92% held out). The basiconic code was nearly one-dimensional (participation ratio 1.71), organized along an axis of response strength with 1-octen-3-ol at its extreme. The CO 2 -sensitive ab4 class was the most sensitive detector for 11 of 16 odorants, whereas ab2 carried no detectable odor information. Classes shared up to 0.41 bits of their information with one another, an overlap only weakly predicted by the similarity of their mean tuning. On a shared odor panel, Drosophila receptors carried twice the information per channel and, with channel number matched, spanned about 1.5-fold more dimensions. These conclusions held in simulations at the real sample sizes, under alternative noise models, and when classes were defined on independent odorants. The tsetse antenna thus combines redundant, detection-oriented basiconic channels with trichoid sensilla that carry host-odor identity, principles relevant to the design of odor baits.

bioRxiv (Cold Spring Harbor Laboratory)
Yale University (US)
Neurobiology and Insect Physiology Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon