Home → Flash News →
ep 175 with jens carlsson clip 3
AI is changing how we think about structure, function, and discovery
Published on
November 4, 2025
Category
AI is changing how we think about structure, function, and discovery — but Jens Carlsson reminds us: the real test isn’t whether we can explain today’s data, it’s whether we can predict tomorrow’s results.
In this clip, Jens unpacks a mindset shift that every computational scientist should hear. For decades, molecular modeling focused on explaining — rationalizing why a ligand binds, why a mutation shifts signaling, why a simulation behaves a certain way. But explaining is retrospective. Prediction is transformative.
With tools like AlphaFold, his lab can now model receptor–peptide complexes in seconds — an achievement that once took months. Yet Jens’s caution stands: prediction must be proven. A good model doesn’t just match published structures; it forecasts new biology, new ligands, and new function.
That’s where the future of GPCR research lies — not in describing the past, but in anticipating it.
Whether you’re running simulations, screening compounds, or designing next-gen molecules, this insight reframes what “success” in computational pharmacology really means.
👉 Watch the full episode: model predict discover
#GPCR #DrGPCR