Ask the author session


Glucagon-like peptide-1 receptor (GLP1R) agonists have transformed clinical care across type 2 diabetes, obesity, and hepatology. However, hepatologists and endocrinologists frequently encounter a clinical puzzle: patient response to GLP-1 therapy is highly individual, with some individuals showing no response at all. Natural genetic variations—specifically missense variants in the GLP1R gene—can alter receptor function and cell signaling, influencing both disease susceptibility and individual drug response.

👉 You can watch the full interview below between Prof. Jana Selent and Prof. Jean-François Dufour — don’t miss it.

The Breakthrough: Bringing Cellular Context to AI

In a recent landmark study published in Diabetologia, Prof. Jana Selent from the University Pompeu Fabra in Barcelona introduced a novel computational framework called GPCRVP (or the GPCR score). While existing artificial intelligence models such as AlphaMissense and REVEL can predict variant effects, they frequently produce false predictions because they evaluate proteins in isolation without their natural cellular environment.

To bridge this gap, Prof. Selent’s team combined machine-learning predictors with all-atom molecular dynamics (MD) computer simulations of GLP1R embedded in a native-like cell membrane.


Key Clinical Takeaways

Please find below the main insights:

  • 78% Fewer False Positives: By factoring in the lipid membrane, allosteric communication, and mutation-induced structural changes, GPCRVP achieved a ~78% reduction in false-positive predictions compared to AlphaMissense alone.
  • Predicting Receptor Function: The score reliably forecasts whether a specific variant impairs crucial receptor functions, such as signal transduction, insulin release, and body weight regulation.
  • Accelerating Precision Medicine: The predictor provides actionable insights into variant impact without requiring time-consuming and expensive laboratory experiments, helping clinicians anticipate patient risk for obesity or type 2 diabetes and avoid trial-and-error prescribing.


Watch the “Ask the Author” Interview

Want to understand how computational biology and membrane-aware AI modeling are shaping the future of personalized metabolic and liver care?

🎥 Watch our full “Ask the Author” interview with Prof. Jana Selent to hear directly from the lead author about the science behind the GPCR score and its potential for precision medicine.

Associated publication: “The GPCRVP score reliably predicts the impact of GLP1R human variants on receptor function”

Thank you to our sponsors