Image
When agricultural weeds survive herbicide applications that would normally control them, resistant populations can emerge and spread. This can lead to increased herbicide use, greater competition with crops for resources, and the need for more complex management strategies, all of which increase costs and pressures on farmers. As a result, managing herbicide resistance has become a critical challenge for both farmers and the wider agricultural sector.
Funded by Innovate UK’s Knowledge Transfer Partnership award, the new collaboration will see genomics and AI expertise from Earlham Institute embedded within Syngenta’s Jealott’s Hill International Research Centre in Berkshire. The project will use genome large-language models - large-scale AI networks trained on genome sequences to predict patterns, structures, and relationships.
Researchers will focus on predicting and understanding target- and non-target site resistance (NTSR), (a challenging form of resistance where weeds survive herbicides by adapting how they absorb, move, or break down the chemical).
Prof Anthony Hall, Head of Plant Genomics at Earlham Institute said “We have a fantastic agri-science community in the UK and this project is a perfect example of applying the latest genomics expertise to an R&D environment, helping develop evidence-based solutions for growers.
“At Earlham Institute we’ve already been developing the tools and methods to implement these models to complex plant biology and believe they have strong potential in unravelling and predicting resistance.”
The project will form part of the Syngenta Crop Protection programme and be led by the Bioscience Digital Group, which applies data science, modelling, and computer vision to develop tools that support early-stage crop protection programmes.
Dr Chris O’Grady, Senior Principal Scientist in Computational Biology at Syngenta said: “We're delighted to be partnering with the Earlham Institute on this Knowledge Transfer Partnership. This project is an exciting opportunity to apply state-of-the-art computational approaches to tackle some of the toughest challenges in weed control.”
Bridging academic and R&D expertise across Earlham Institute and Syngenta, the Knowledge Transfer Partnership will develop computational pipelines and gLLMs. By studying and predicting this type of resistance, the team hopes to develop methods to slow resistance spread, preserve herbicide efficacy, and ultimately minimise excessive chemical use.
Dr. Deepmala Sehgal, Senior Principal Scientist in Herbicide Resistance team at Syngenta, explains: “Existing genetic data from weed species could be fed into AI models to help predict how weeds develop resistance to herbicides “
