Beyond the Hype: The Realities of End-to-End Data-Driven Chemistry in Agrochemical R&D
Crop protection remains critical to global food security, with arthropod pests, fungal pathogens, and competing weeds together accounting for an estimated 20–40% of annual yield losses worldwide1 - figures projected to worsen under climate change. Developing selective, effective, and sustainable solutions to control fungi, insects and weeds is therefore central to Syngenta's mission and drives the need for innovative approaches to accelerate the discovery of novel crop protection active ingredients.
New Artificial Intelligence-based approaches2,3 have garnered significant interest in how they can be integrated into commercial agrochemical discovery pipelines. In this talk, we describe Syngenta's journey in generative chemistry. Since pioneering its use in agrochemistry in 2019, we have accumulated substantial practical experience, and we discuss the realities of deploying generative chemistry in an industrial context: from the design of multiparameter scoring functions (e.g., biological activity, safety, sustainability, and IP novelty), to steering generative algorithms toward desired regions of chemical space, to efficient post-processing strategies.
Yet designing a molecule solves only half the problem - a promising candidate must also be synthesizable. By integrating retrosynthetic prediction4,5 and data mining6, we identify optimal synthetic routes in silico, transforming virtual designs into tangible chemistry. The convergence of generative design and synthesis planning compresses discovery timelines. Today, generative chemistry supports majority of our research projects.
[1] Serge Savary, Laetitia Willocquet, Sarah J. Pethybridge, Paul Esker, Neil McRoberts, Andy Nelson, Nature Ecology & Evolution, 2019, 3, 430-439.
[2] Josh Abramson, Jonas Adler, Jack Dunger, Richard Evans, Tim Green, Alexander Pritzel, Olaf Ronneberger, et al., Nature, 2024, 630, 493-500.
[3] Saro Passaro, Gabriele Corso, Jeremy Wohlwend, Mateo Reveiz, Stephan Thaler, Vignesh Ram Somnath, Noah Getz, et al., BioRxiv, 2025.
[4] Yinjie Jiang, Yemin Yu, Ming Kong, Yu Mei, Luotian Yuan, Zhengxing Huang, Kun Kuang, et al., Engineering, 2023, 25, 32-50.
[5] Marta Pasquini, Marco Stenta, Journal of Cheminformatics, 2023, 15, 41.
[6] Nataliya Lopanitsyna, Marta Pasquini, Marco Stenta, Journal of Cheminformatics, 2026, 18, 6.