Physics, Free Energy, & Drug Discovery: Inside Schrödinger's Computational Platform

About this episode

This episode explores how physics-based computational modeling is transforming drug discovery with insights from Robert Abel of Schrödinger.

In this episode of Data in Biotech, Ross Katz sits down with Robert Abel, Chief Scientific Officer of the Platform at Schrödinger, to explore how physics-based computational modeling is transforming drug discovery. Robert unpacks why machine learning alone isn't enough to navigate the vast complexity of chemical space - an estimated at 10⁶⁰ possible drug-like molecules - and how integrating atomistic simulations with ML creates a more accurate, reliable, and scalable approach to identifying viable drug candidates. From free energy perturbation calculations to generative AI, Robert offers a rare inside look at how Schrödinger's technology platform is accelerating the path from target identification to clinical candidate and where the field is headed next. What you’ll learn in this episode: >> Why chemical space (~10⁶⁰ molecules) makes purely data-driven ML approaches fundamentally insufficient for drug discovery, and how physics-based sampling solves the training data problem >> How free energy perturbation (FEP) calculations enable quantitative prediction of protein-ligand binding affinities at near-experimental accuracy (~1.2 kcal/mol RMSE) >> How Schrödinger's…

People in this episode

Host: Ross Katz

Guest: Robert Abel

Topics covered

• drug discovery
• computational modeling
• machine learning
• chemical space
• free energy perturbation
• AI in biotech

Keywords

• drug-like molecules
• machine learning
• atomistic simulations
• binding affinities
• active learning
• clinical candidate

Mentioned in this episode

Organizations: Schrödinger

More episodes of Data in Biotech

• From Tissue to Mechanism to Decision: Building AI for Computational Oncology · June 2, 2026 · 47 min
• Cavities in the Data: Building FDA-Cleared AI for Dental Imaging with Overjet · May 13, 2026 · 59 min
• Data as a Moat: Why Biotech's Most Valuable Asset is Buried in a Hard Drive · April 30, 2026 · 43 min
• Markus Gershater on Why Experimental Design in Biotech is Broken and How to Fix It · April 15, 2026 · 41 min
• Data Science and Diagnostic Models - the What, Why and How with Michelle Wiest · April 15, 2026 · 44 min
• The Patient is Not a Document: Foundation Models for Biomedical AI with Standard BioModel · April 15, 2026 · 50 min