The Power of AI Applied to Agrochemical Bioactivity

In the face of growing agrochemical resistance and increasingly stringent regulatory requirements, how can artificial intelligence (AI) be harnessed to help lower the costs, failure rates and timelines associated with current agrochemical development cycles?

Drawing on case study evidence, we explore how deep learning techniques allow us to overcome many of these challenges.

Our presenters, George Lahm and Laurie Christianson, from FMC, along with Bailey Montefiore and Matt Segall, from Optibrium, share findings from their collaborative project on the application of the Alchemite™ deep learning imputation method, used in Optibrium’s Cerella™ platform, to compounds in the early stages of discovery.

Talking points will include how deep learning methods can:
• More accurately predict complex experimental endpoints measured in whole organisms (plants and insects) that are challenging for predictive models
• Save time and resources by focusing on the most valuable experiments to predict activities across a broad range of weed and pest species
• Identify the best overall compounds based on limited information – and inform new compound selection

_{To hear the narration, please increase your speaker volume.}