Researchers at the Oregon State University College of Engineering have used the power of artificial intelligence to help protect bees from pesticides.
Cory Simon, assistant professor of chemical engineering, and Xiaoli Fern, associate professor of computer science, led the project, which included training in a machine learning model to predict whether any new herbicide, fungicide or insecticide would be proposed. can be toxic to honey bees based on the molecular structure of the compound.
The findings, shown on the cover of The Journal of Chemical Physics in a special issue, “Chemical Design by Artificial Intelligence,” is important because many fruit, nut, vegetable and seed crops rely on bee pollination.
Without bees to transfer the pollen needed for breeding, nearly 100 commercial plants in the United States would disappear. The global economic impact of bees each year is estimated to exceed $ 100 billion.
“Pesticides are widely used in agriculture, increasing crop yields and providing food security, but pesticides can harm non -target species such as bees,” Simon said. “And because insects, weeds, etc. eventually develop resistance, new pesticides must continue to be developed, without harming the bees.”
Graduate students Ping Yang and Adrian Henle used honey bee toxicity data from pesticide exposure experiments, involving nearly 400 different pesticide molecules, to train an algorithm to predict whether a new pesticide molecule can be toxic to honey bees.
“The model represents pesticide molecules by a set of random walks on their molecular graphs,” Yang said.
A random walk is a mathematical concept that describes any winding path, such as the complex chemical structure of a pesticide, where each step of the path is chosen randomly, as by tossing coins. .
Imagine, Yang explains, that you are walking aimlessly along the chemical structure of the pesticide, moving from atom to atom through the bonds that hold the compound together. You travel in random directions but keep an eye on your route, the sequence of atoms and bonds you visit. Then you come out with another molecule, comparing a series of twists and turns to what you’ve done before.
“The algorithm declares two molecules to be the same if they have multiple motions with the same sequence of atoms and bonds,” Yang said. “Our model serves as a surrogate for a bee toxicity experiment and can be used to quickly screen proposed pesticide molecules for their toxicity.”
This research was supported by the National Science Foundation.
Materials provided by Oregon State University. Originally written by Steve Lundeberg. Note: Content can be edited for style and length.