The abstract is read at the publisher; the summary is JClub's.
A machine-learning pipeline predicts and validates repellent odorants for honey bees (Apis mellifera), identifying >130 candidates and confirming strong field repellency for seven top chemicals.
- Why it matters: Finding odorants that deter beneficial pollinators is hard because honey bee olfaction involves >170 receptors that map volatiles to valence, limiting discovery from complex biology.
- What they did: The team trained a structure-to-aversive-valence model on published behavior, refined it with new species-level data for Apis mellifera and Drosophila, then screened >50 million compounds to generate >130 repellent candidates.
- The result: Laboratory behavior testing showed high predictive success, and field assays with freely foraging Apis mellifera confirmed strong repellency for the top seven candidates, implying high likelihood of repelling foragers from pesticide-treated crops.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.