CONTACT
LOG IN | REGISTER

Contribution toward safer and sustainable- by-design crop protection products

David Rouquie

ECETOC, Belgium

Bio

I lead the Toxicology Data Science team at the Bayer Crop Science toxicology facility in Sophia Antipolis, France.

My passion for science and research that addresses societal needs has kept me deeply engaged in innovative, collaborative, and multidisciplinary projects. With a background in biochemistry and molecular biology, I am continually evolving into a hybrid profile at the intersection of biological and computational sciences.

I serve as Bayer’s representative and vice Chairman of the ECETOC scientific committee. In the context of the Broad Institute/HESI collaboration, I co-chair the OASIS initiative. This initiative aims to build confidence in the use of Cell Painting, transcriptomics, and proteomics coupled with exposure modeling for chemical safety assessments, using hepatotoxicity as a case study.

Abstract

Contribution toward safer and sustainable- by-design crop protection products

Advances in artificial intelligence (AI) and high-content screening technologies is revolutionizing the fields of small molecule discovery and toxicology.

In this presentation, we explore the integration of AI with cell painting data—a morphological profiling method that captures cellular responses to chemical exposure—to enhance toxicity prediction and facilitate de novo compound design. By employing machine learning algorithms, we extract complex patterns from high-dimensional cell painting images, enabling more accurate identification of toxicity profiles and biological mechanisms triggered by chemical compounds.

Furthermore, as proof of concept we apply AI-guided generative models to design novel compounds with desired biological properties. This approach not only accelerates the identification of safe and effective candidates but also reduces reliance on costly and time-intensive animal testing. Our findings demonstrate that the synergy between AI and cell painting data holds significant promise for predictive toxicology and chemical design, paving the way for more efficient and data-driven approaches in chemical safety assessment and small molecule discovery.

crosschevron-down