Francesca Grisoni is an Assistant Professor in the Department of Biomedical Engineering, where she leads the Molecular Machine Learning team. Her research focuses on developing novel artificial intelligence (AI) methods for drug discovery, at the interface between computation and experimental validation in the wet-lab. Her team works at the intersection of (and melds concepts from) chemistry, biology, and computer science, to design innovative AI approaches tailored to molecule discovery. Her final goal is to augment human intelligence with artificial intelligence, to ultimately reach ‘better decisions faster’ in the discovery of therapeutic drugs.
My mission is to stretch the limits of AI in drug discovery, in order to create cutting-edge technology able to augment human creativity in the discovery of next-generation therapeutics.
Francesca Grisoni studied Environmental Sciences at the University of Milano-Bicocca. There she also obtained her Ph.D. in 2016, working under the supervision of Prof. R. Todeschini, with a dissertation on interpretable machine learning for molecular property prediction. During her doctoral studies, she was hosted at ETH Zurich (Dept. Chemistry and Applied Biosciences) and at the U.S. EPA (National Center of Computational Toxicology). After working as a biostatistical consultant for Bracco Pharmaceuticals, and as a data scientist in a startup company for a year, Dr. Grisoni returned to academia. From 2017 to 2019, she was a joint postdoctoral fellow at the University of Milano-Bicocca and ETH Zurich, where she developed novel molecular descriptors tailored to scaffold hopping. In 2019 she joined the group of Prof. Gisbert Schneider at ETH Zurich as a senior postdoctoral researcher, where she focused on generative deep learning for drug discovery. In 2021, she was appointed Assistant Professor at the TU/e, where she currently leads the Molecular Machine Learning team. Dr. Grisoni has received several grants and awards, such as the Lush Young Researcher Prize, the Early Career Award 2022 from the Dutch Royal Netherlands Academy of Arts and Sciences (KNAW), and an ERC Starting Grant (2022).
Chemical language models for de novo drug designCurrent Opinion in Structural Biology (2023)
Leveraging molecular structure and bioactivity with chemical language models for de novo drug designNature Communications (2023)
Structure-based drug discovery with deep learningarXiv (2022)
Exposing the Limitations of Molecular Machine Learning with Activity CliffsJournal of Chemical Information and Modeling (2022)
Perplexity-Based Molecule Ranking and Bias Estimation of Chemical Language ModelsJournal of Chemical Information and Modeling (2022)
Current Educational Activities
No ancillary activities