Federated Learning in Drug Discovery: How Lilly TuneLab Integrates with CDD Vault
Thursday October 22, 2026 9 AM (PT) | 12 PM (ET) | 6 PM (CEST)
Reserve Your Webinar SeatJoin Eli Lilly's Jonathan Gilbert, Ph.D. and CDD's Peter Gedeck, Ph.D. for a technical webinar showcasing Lilly TuneLab’s predictive AI/ML models embedded directly within CDD Vault. The session will first introduce the fundamentals of federated learning and explain how TuneLab uses this approach to build drug discovery models trained on decades of Lilly proprietary research data. We will discuss ChemLab which predicts small molecule ADMET endpoints and AbLab which predicts antibody developability endpoints. We will then walk through the specific integration, showing how TuneLab models fit within the CDD Vault environment, providing model access where discovery work already happens.
What You’ll Learn
-
How TuneLab brings Lilly-trained predictive AI/ML models directly into CDD Vault workflows
-
How federated learning enables models to benefit from decades of Lilly research data
-
How ChemLab supports small-molecule ADMET predictions
-
How AbLab supports antibody developability predictions
-
How scientists can apply these models within their existing drug discovery workflows
Speakers
Jonathan Gilbert, PhD
Senior Director – Ecosystem Growth & Contributor Partnerships, Eli Lilly
Jonathan Gilbert leads the growth of the Lilly TuneLab Ecosystem. Jonathan received his PhD from MIT in Chemical Engineering and has spent his last decade plus in biotech at multiple companies ranging from the Seed stage to early clinical public companies. Immediately prior to Lilly, Jonathan led Corporate Development and Strategy at Entact Bio.
Peter Gedeck, PhD
Research Informatics Senior Scientist, Collaborative Drug Discovery (CDD Vault)
Peter Gedeck has over 30 years of experience in scientific computing and data science. After 20 years as a computational chemist at Novartis, he now works as a senior data scientist at Collaborative Drug Discovery. He specializes in the development of machine learning algorithms to predict biological and physicochemical properties of drug candidates.

