Digital Twin Maps Tumor Metabolism in Glioma
Researchers at the University of Michigan published in Cell Metabolism (2025) a machine-learning "digital twin" that maps real-time tumor metabolism in glioma patients to predict metabolic flux. Trained on synthetic data constrained by labeled-glucose infusions from eight patients and validated against six held-out patients plus mouse experiments, it identifies tumors likely to respond to amino-acid–restricted diets or mycophenolate mofetil, enabling virtual testing of personalized metabolic therapies.
Key Points
- 1Uses CNN digital twin trained on synthetic and labeled-glucose infusion data from eight glioma patients
- 2Detects tumors' capacity to synthesize amino acids, distinguishing likely responders to dietary metabolic therapies
- 3Allows clinicians to virtually test diets and drugs, reducing ineffective treatments and personalizing therapy choices
Scoring Rationale
High novelty with peer-reviewed validation and preclinical confirmation, but limited patient sample and early-stage translational readiness.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
