Machine learning models enabling early and high-throughput prediction of ADME properties and pharmacokinetics
Includes a Live Web Event on 08/25/2026 at 3:00 PM (EDT)
-
Register
- Non-member - $49
- Member - Free!
- Student - Free!
- Premier - Free!
This webinar is designed to provide an overview of emerging strategies transforming pharmacokinetic (PK) prediction in modern drug discovery. It will explore how high-throughput physiologically based pharmacokinetic (HT-PBPK) modelling, combined with machine learning, enables rapid, structure-driven prediction of PK properties, reducing reliance on experimental data in early discovery.
This session will highlight the application of HT-PBPK workflows for large-scale compound prioritisation, alongside advances in modelling beyond Rule-of-Five (bRo5) compounds such as PROTACs and cyclic peptides. Through case studies, speakers will demonstrate how novel physicochemical descriptors improve the prediction of key ADME and in vivo PK endpoints, including permeability, solubility, and systemic exposure. Together, these approaches offer new opportunities to enhance decision-making and accelerate the design of developable drug candidates.
By the end of this webinar, participants will be able to:
1. Describe how in silico inputs can replace in vitro data to enable scalable pharmacokinetics prediction. 2. Assess the value of novel physicochemical descriptors in improving ADME and pharmacokinetics predictions
3. Interpret prediction accuracy (e.g., fold-error ranges) and identify key success factors (e.g., clearance pathway classification, ML confidence).
4. Understand the drug development challenges posed by PROTACs and cyclic peptides
5. Describe emerging strategies for modelling beyond Rule-of-Five (bRo5) compounds
Davide Bassani
Computational DMPK and Translational PK/PD Leader
Hoffmann-La Roche
Computational DMPK Leader and Translational PK/PD Project Leader at Roche in Basel, specializing in the application of advanced computational methods, in silico DMPK/PD, and translational pharmacology to accelerate small-molecule drug discovery. With a Ph.D. in Computer-Aided Drug Design, I leverage machine learning, in silico PK prediction, and medicinal chemistry expertise to guide portfolio projects and strengthen scientific decision-making. My work focuses on optimizing screening strategies, improving cycle times, and integrating ADME, DMPK, and pharmacodynamic insights to advance promising candidates toward the clinic.
Jeremy Jones
Principal Scientist
Simulations Plus
Dr. Jones received his PhD from Stanford University and completed his postdoctoral fellowship at UC San Francisco, before becoming an Assistant Professor in Molecular Pharmacology at City of Hope. There he focused on drug development and translational research in urologic oncology, developing and licensing a lead series for prostate cancer. He also spent a year at Terray Therapeutics before moving to New Zealand, and now has been working at Simulations Plus on the early drug discovery team for over three years.
Daniel Scotcher (Moderator)
Lecturer
The University of Manchester
Dr. Dan Scotcher is a Lecturer in Applied Pharmacokinetics at the University of Manchester, where his research focuses on mechanistic modelling and simulation to understand drug disposition and optimise drug dosing. He has extensive experience in physiologically based pharmacokinetic (PBPK) modelling and its application across drug development and clinical pharmacology. Dan currently serves as Chair of the ISSX Modelling & Simulation (M&S) Focus Group.
William Weigel (Moderator)
Scientist, Molecular Design
Terray Therapeutics
At Terray Therapeutics, my role involves the design and analysis of small-molecule combinatorial libraries used to drive hit finding around protein targets of interest. My academic background was focused on synthetic organic chemistry during my PhD and work with DNA Encoded Libraries (DELs) during my postdoc.
Key:
Complete
Next
Failed
Available
Locked