About the Talk

Presenter

Francesco Gentile: Assistant Professor, University of Ottawa

Francesco Gentile is an Assistant Professor in the Department of Chemistry and Biomolecular Sciences at the University of Ottawa and a Principal Investigator at the Ottawa Institute of Systems Biology. He holds an MSc in Bioengineering from the Politecnico di Torino and a PhD in Biophysics from the University of Alberta, and completed a CIHR/MSFHR-funded postdoctoral fellowship at the University of British Columbia before joining the University of Ottawa in 2022. His group develops computational and AI methods for the discovery of small molecules and biotherapeutics across cancer, neurological diseases and antimicrobial resistance. He is a co-founder of InVirtuo Laboratories, an AI-driven drug discovery company based in Lugano, Switzerland.

Abstract

Artificial intelligence, particularly machine learning (ML), is playing an increasingly important role in the discovery of synthetic molecules that bind to proteins and nucleic acids, modulating their activity for therapeutic purposes. However, the effectiveness of ML methods is often constrained by the need for large volumes of high-quality data, limiting their utility in data-scarce scenarios. In contrast, physics-based approaches can generate large, consistent, and reproducible datasets, and their integration with ML can significantly enhance performance and generalizability. In this talk, I will present our recent efforts to develop hybrid ligand discovery methods that combine the strengths of ML and physics-based simulations to accelerate and improve the drug discovery process. I will first describe how we leverage ML-accelerated simulations to explore chemical space more efficiently, showcasing applications in the discovery of ligands for the SARS-CoV-2 non-structural protein 3, inhibitors of amyloid beta-42 aggregation, and modulators of key antibiotic resistance proteins. I will also highlight novel ML approaches developed in our lab to improve native pose recognition and binding affinity prediction, with a focus on maintaining broad generalizability. Together, these efforts underscore the enduring value of computational chemistry at the dawn of the AI era and its pivotal role in shaping the future of ligand discovery.

Date: July 9th, 2025 – 5:00 pm (GMT+3)

Language: English