Related Experiment Video
Updated: Nov 5, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.1K
Aging-related markers in rat urine revealed by dynamic metabolic profiling using machine learning
Dan Shi1,2, Qilong Tan3, Jingqi Ruan1
1National Key Discipline Laboratory, Department of Nutrition and Food Hygiene, School of Public Health, Harbin Medical University, Harbin, PR China.
Aging
|May 21, 2021
Summary
Researchers identified six key metabolites in rat urine that accurately track aging. These novel metabolic biomarkers could be crucial for developing therapies to delay the aging process and improve healthspan.
Area of Science:
- Metabolomics
- Gerontology
- Biomarker Discovery
Background:
- Aging and metabolism are closely linked, necessitating the development of metabolic biomarkers to understand and potentially delay aging.
- Current machine learning approaches have limited reliable markers for aging trajectories.
Purpose of the Study:
- To identify reliable metabolic biomarkers reflecting aging trajectories using machine learning.
- To discover novel therapeutic targets for antiaging interventions.
Main Methods:
- Generated urine metabolomic profiles from rats at four age stages (20, 50, 75, 100 weeks) using ultra-performance liquid chromatography/mass spectrometry.
- Applied four algorithms (VIP, time-series, LASSO, SVM-RFE) to screen aging-related biomarkers.
- Validated identified biomarkers in an independent test group.
Main Results:
- Partial least squares-discriminant analysis showed clear age-related separation in metabolic profiles.
- Successfully screened 25 aging-related biomarkers and validated them with perfect accuracy (AUC=1).
- Identified six key metabolites: epinephrine, glutarylcarnitine, L-kynurenine, taurine, 3-hydroxydodecanedioic acid, and N-acetylcitrulline as aging markers.
Conclusions:
- The study reveals distinct metabolic trajectories associated with aging in rats.
- The identified metabolites serve as novel, reliable biomarkers for aging.
- These biomarkers represent potential therapeutic targets for antiaging strategies.
