OSU researchers use artificial intelligence to save bees from pesticides
Everyone says, 'save the bees,' but researchers at Oregon State University College of Engineering have developed artificial intelligence to do just that.
The project, headed by assistant professor of chemical engineering Cory Simon and associate professor of computer science Xiaoli Fern, entailed using a machine learning model to predict the toxicity of new herbicides, insecticides or fungicides toward bees through their molecular structures. The National Science Foundation supported this research.
The results, published in The Journal of Chemical Physics' special issue "Chemical Design by Artificial Intelligence," are significant due to the dependence of many if not most fruit, vegetable, seed and nut crops on bee pollination.
If bees disappeared, so would almost 100 commercial crops in the United States. Additionally, bees' annual global economic contribution is estimated to surpass $100 billion.
"Pesticides are widely used in agriculture, which increase crop yield and provide food security, but pesticides can harm off-target species like bees," Simon said. "And since insects, weed, etc. eventually evolve resistance, new pesticides must continually be developed, ones that don't harm bees."
Graduate students Ping Yang and Adrian Henle fed the artificial intelligence honeybee toxicity data from pesticide exposure experiments to predict if new pesticide molecules would be toxic to bees.
"The model represents pesticide molecules by the set of random walks on their molecular graphs," Yang said.
A random walk is a math concept which predicts what a path, in this case a path along the chemical structure of a pesticide, will look like if left up to random chance.
"Imagine, Yang explains, that you're out for an aimless stroll along a pesticide's chemical structure, making your way from atom to atom via the bonds that hold the compound together," a release from OSU said. "You travel in random directions but keep track of your route, the sequence of atoms and bonds that you visit. Then you go out on a different molecule, comparing the series of twists and turns to what you've done before."
"The algorithm declares two molecules similar if they share many walks with the same sequence of atoms and bonds. Our model serves as a surrogate for a bee toxicity experiment and can be used to quickly screen proposed pesticide molecules for their toxicity."
You count on us to stay informed and we depend on you to fund our efforts. Quality local journalism takes time and money. Please support us to protect the future of community journalism.