Scientists use AI to create bee repellent, keep bees away from pesticides
California: In a win for pollinators, researchers at UC Riverside have found a way to keep honey bees away from pesticide-treated crops — using smell.
Pesticide exposure is a major driver of bee decline and colony collapse. The challenge: bees have over 200 odor receptors, making it extremely hard to find a scent that repels rather than attracts them.
Biologist Anandasankar Ray and entomologist Boris Baer solved it with AI. They built a machine-learning model trained on chemical structures and bee behavioral responses, refined it with lab tests on honey bees and fruit flies, then screened over 50 million compounds.
The model flagged 130 promising candidates. Seven top compounds were tested, and all of them caused clear avoidance in lab and field trials — free-foraging bees were reliably repelled from honeycombs without being harmed.”By keeping bees away from harmful pesticides, we can potentially reduce their risk of exposure without compromising the protection of crops,” Ray said.
The compounds, now part of a patent application published in eLife, could lead to bee-friendly pesticide formulations that include a built-in repellent.
Beyond farms, the repellents could also reduce human-bee conflicts in hospitals and residential areas, or prevent pollination in seedless fruit crops.”Protecting pollinators doesn’t have to come at the expense of food security,” Ray said.
