Researchers at the University of California, Davis, employed artificial intelligence to enhance plants’ ability to identify a broader spectrum of bacterial dangers, perhaps resulting in novel methods to safeguard crops such as tomatoes and potatoes against severe diseases. The research was published in Nature Plants.

Plants possess immune systems analogous to those of animals. Their defence arsenal include immunological receptors that enable the detection and defence against germs. One receptor, known as FLS2, assists plants in identifying flagellin – a protein found in the diminutive appendages that bacteria utilise for locomotion. Bacteria are cunning and always changing to evade detection.

“Bacteria are in an arms race with their plant hosts, and they can change the underlying amino acids in flagellin to evade detection,” stated lead author Gitta Coaker, a professor in the Department of Plant Pathology.

To enhance plant resilience, Coaker’s team employed natural variation alongside artificial intelligence—specifically AlphaFold, a technique designed to predict the three-dimensional structure of proteins, and reengineered FLS2, thereby augmenting its immune system to detect additional intruders.

The scientists concentrated on receptors already identified to recognise a greater number of bacteria, regardless of their absence in beneficial crop species. Through a comparative analysis with more specialised receptors, the researchers identified the specific amino acids to modify.

“We were able to resurrect a defeated receptor, one where the pathogen has won, and enable the plant to have a chance to resist infection in a much more targeted and precise way,” Coaker stated.

Significance

Coaker stated that this facilitates the advancement of broad-spectrum disease resistance in crops through predictive design.

A primary objective of the researchers is a significant agricultural threat: Ralstonia solanacearum, the pathogen responsible for bacterial wilt. Certain strains of the soil-borne virus can infect over 200 plant species, including essential crops such as tomato and potato.

The team is using machine learning methods to forecast which immune receptors merit altering in the future. They are also attempting to reduce the quantity of amino acids that require modification.

This methodology may enhance the perceptual capacity of additional immune receptors using a comparable mechanism.

The study also includes authors Tianrun Li, Esteban Jarquin Bolaños, Danielle M. Stevens, and Hanxu Sha from UC Davis, as well as Daniil M. Prigozhin from Lawrence Berkeley National Laboratory.

The study received funding from the National Institutes of Health and the National Institute for Food and Agriculture of the United States Department of Agriculture.

Original Publication
Authors: Tianrun Li, Esteban Jarquin Bolaños, Danielle M. Stevens, Hanxu Sha, Daniil M. Prigozhin and Gitta Coaker.
Journal: Nature Plants
DOI: 10.1038/s41477-025-02049-y
Method of Research: Experimental study
Article Title: Unlocking expanded flagellin perception through rational receptor engineering
Article Publication Date: 28-Jul-2025