Wake up and smell the infected bees: Volatile cues of Vairimorpha infection in honey bees
Data files
Jun 05, 2026 version files 79.12 KB
-
Analysis_script.R
19.22 KB
-
experimental_spores.csv
256 B
-
metadata.csv
4.11 KB
-
README.md
2.44 KB
-
reference_spores.csv
165 B
-
VOC_normalised_area.csv
49.86 KB
-
VOC_table.csv
3.08 KB
Abstract
Infection can alter volatile organic compounds (VOCs) across many taxa, yet responses in adult social insects remain poorly characterised. We quantified VOC emissions from worker honey bees experimentally infected with Vairimorpha spp. and from uninfected controls across a 14-day time series following inoculation. Headspace volatiles were collected using dynamic headspace sampling and analysed using comprehensive two-dimensional gas chromatography coupled with mass spectrometry (GC×GC-MS).
VOC profiles differed between infected and control bees when pooled across sampling days, with particularly strong divergence at 6 and 12 days post-infection. Consensus between multivariate analyses, including random forest classification and redundancy analysis, identified tetradecane, dodecane and one unidentified compound as characteristic of early infection, whereas 3,4-dimethylbenzaldehyde was associated with later infection stages.
Our findings document temporal shifts in the chemical phenotype of infected workers and provide a resource for comparative analyses of odour variation, host–pathogen interactions and disease detection in social insects.
Dataset DOI: 10.5061/dryad.ttdz08mc3
Description of the data and file structure
Each VOC profile is linked to accompanying metadata describing treatment, days post-infection, colony of origin, and replicate identity.
Files and variables
File: Analysis_script.R
File: experimental_spores.csv
Description: Pooled spore loads of bees from each cage of bees following headspace sampling 14 days post-infection with Vairimorpha spp.
Variables
- cage_ID: Identity of cage used for VOC sampling
- hive: Colony of origin
- treatment: Treatment applied
- dpi: Days post-infection
- spores/bee: Vairimorpha spores per bee
File: metadata.csv
Description: Metadata accompanying each sample
Variables
- cage_ID: Identity of cage used for VOC sampling
- hive: Colony of origin
- treatment: Treatment applied
- spore_load: Spore load of inoculum
- tube_no: Thermal desorption tube identifier
- date: Date sampled
- dpi: Days post-infection
- no_bees_sampled: Number of bees sampled
- Treat: Short-hand treatment identifier
- Sample Name: Unique identifier combining Cage ID, Treatment, and time period
- Tube No: Second thermal desorption tube identifier for ensuring correct sample alignment
- Values 'BLANK' refer to samples of background emissions used as blanks during processing
- Values '-' indicate not applicable
File: reference_spores.csv
Description: Timeline of spore load progression following Vairimorpha infection in reference bees to track spore progression over 14 days post-infection
Variables
- hive: Colony of origin
- dpi: Days post-infection
- spores/bee: Vairimorpha spores per bee
File: VOC_table.csv
Description: Compound name, component, CAS registry number and chemical class of each VOC detected.
Variables
- compound: Compound name
- component: Component to align with VOC abundance data
- cas.number: CAS registry number
- family: Chemical class
File: VOC_normalised_area.csv
Description: Normalised VOC abundance/area from GCxGC-MS
Variables
- Sample: Unique identifier combining Cage ID, Treatment, and time period
- C001:C071: Individual VOCs (components) identified in GCxGC-MS analysis.
Code/software
All analyses were conducted in R v4.5.2
Data were generated from an experimental infection study using adult worker honey bees (Apis mellifera) exposed to Vairimorpha spp. or maintained as sham-treated controls. Individual bees were sampled at multiple time points over a 14-day post-infection period. Volatile organic compounds (VOCs) were collected from groups of live caged bees using dynamic headspace sampling and quantified using comprehensive two-dimensional gas chromatography coupled with mass spectrometry (GC×GC-MS).