Data-driven or physics-based: Choosing the smarter approach to DMPK modeling

Explore the strengths and challenges of different approaches to DMPK modeling in this on-demand webinar, where data science meets drug development.

This session takes a close look at both data-driven models, which identify patterns from existing datasets, and physics-based models, which rely on fundamental equations.

You'll learn about common data limitations, including heterogeneity and bias, and why understanding chemical space and domain applicability is essential for building accurate, reliable predictions.

The webinar also covers key topics like effective data management, various data splitting techniques and their impact on model performance, and emerging AI applications shaping the future of DMPK.

Key topics covered:

• Foundations and limitations of DMPK modeling
• Data curation and AI model evaluation in DMPK
• What’s next: mechanisms, interpretability, and the role of physics-based approaches in future models

About the speakers

• Bart Lenselink - Director, Lead Cheminformatics and Data Science Technology at Structure Therapeutics
• Daniel Price - Vice President, Computational Chemistry and Structural Biology at Nimbus Therapeutics
• John Maclean - Senior Principal Scientist, Computational Chemistry and Informatics at Pharmaron
• Moderated by Nicolas Duchemin - Business Development Manager at Pharmaron