Related Experiment Video
Updated: Aug 10, 2025
06:13
Author Spotlight: Exploring the Impact of Reduced Resistance Exercise Volume on Metabolic Health
Published on: December 1, 2023
1.2K
A machine-learning algorithm integrating baseline serum proteomic signatures predicts exercise responsiveness in
Candela Diaz-Canestro1, Jiarui Chen1, Yan Liu1
1State Key Laboratory of Pharmaceutical Biotechnology, The University of Hong Kong, Hong Kong, China; Department of Medicine, The University of Hong Kong, Hong Kong, China.
Cell Reports. Medicine
|February 14, 2023
Summary
High-intensity exercise training alters hundreds of proteins linked to metabolism and inflammation in men with prediabetes. A machine-learning model predicts individual responses to exercise, aiding diabetes prevention.
Area of Science:
- Exercise physiology
- Metabolomics
- Proteomics
Background:
- Chronic exercise benefits diabetes prevention, but underlying molecular mechanisms are not fully understood.
- Prediabetes increases risk for type 2 diabetes, necessitating effective prevention strategies.
Purpose of the Study:
- To investigate molecular changes in serum proteins associated with exercise in prediabetes.
- To identify predictors of metabolic response to exercise training.
Main Methods:
- Serum proteomic profiling of 688 biomarkers in 36 men with prediabetes.
- 12-week high-intensity interval exercise training intervention.
- Machine-learning analysis to predict exercise response.
Main Results:
- Hundreds of exercise-responsive proteins were identified, regulating metabolism, cardiovascular function, inflammation, and apoptosis.
- Proteins involved in gastrointestinal mucosal immunity showed strong associations with metabolic outcomes.
- Exercise-induced changes in trefoil factor 2 (TFF2) correlated with insulin resistance.
- Baseline levels of glycoprotein 2 (GP2) predicted changes in glucose tolerance.
- A 23-protein signature, including TFF2, differentiated exercise responders from non-responders.
- A machine-learning model accurately predicted metabolic responsiveness using baseline proteomic data.
Conclusions:
- High-intensity interval exercise training induces significant proteomic alterations in men with prediabetes.
- Specific proteins and baseline proteomic signatures can predict metabolic response to exercise.
- These findings offer insights into molecular pathways for exercise-based diabetes prevention and personalized interventions.
