Related Experiment Video
Updated: May 14, 2026
09:16
Real-Time Analysis of Bioenergetics in Primary Human Retinal Pigment Epithelial Cells Using High-Resolution Respirometry
Published on: February 3, 2023
Interpretable machine learning with SHAP analysis identifies redox-modulating dietary antioxidants for predicting
Bei Zhang1, Xinyu Zhang1, Lanyue Ma1
1The fourth Clinical Medical College of Guangzhou University of Chinese Medicine, Shenzhen, Guangdong, China.
Experimental Gerontology
|May 12, 2026
Summary
This study identifies key dietary antioxidants, including daidzein, apigenin, magnesium, zinc, and vitamin E, that are linked to reduced accelerated biological aging risk. Machine learning models accurately predict aging status based on these nutritional factors.
Area of Science:
- Nutritional Science
- Gerontology
- Computational Biology
Background:
- Biological aging, a decline in physiological function, is better measured by biological age than chronological age.
- Dietary antioxidants are crucial for delaying aging, but identifying key components and their complex interactions in diet is challenging.
- Existing statistical methods struggle with high-dimensional dietary data, necessitating advanced analytical approaches.
Purpose of the Study:
- To develop an interpretable machine learning model for predicting accelerated biological aging.
- To identify core dietary antioxidants associated with accelerated aging risk.
- To provide a population-level reference for nutritional interventions promoting healthy aging.
Main Methods:
- Analysis of U.S. National Health and Nutrition Examination Survey (NHANES) data (2007-2010, 2017-2018) from 8125 participants.
- Definition of accelerated biological aging using Klemera-Doubal Method (KDM) and Phenotypic Age (PhenoAge) metrics.
- Application of XGBoost and SHapley Additive Explanations (SHAP) for model construction, performance evaluation (10-fold cross-validation), and feature importance analysis.
Main Results:
- XGBoost model achieved high predictive performance for accelerated aging (AUC 0.931 for KDM, 0.906 for PhenoAge).
- SHAP analysis identified daidzein, apigenin, magnesium, zinc, and vitamin E as key dietary antioxidants predicting accelerated aging risk.
- Higher intake of these five antioxidants showed significant inverse associations with accelerated aging risk across subgroups.
Conclusions:
- An interpretable machine learning model effectively predicts accelerated biological aging based on dietary antioxidant profiles.
- Daidzein, apigenin, magnesium, zinc, and vitamin E are identified as core dietary antioxidants associated with reduced accelerated aging risk.
- Findings support nutritional strategies for healthy aging but require validation in prospective studies due to the cross-sectional design.
Related Concept Videos
01:26Aging
Aging is a complex biological phenomenon influenced by various processes that affect cellular and systemic functions. Several prominent theories attempt to explain its mechanisms, highlighting cellular limitations, oxidative damage, and hormonal changes as central factors in aging.
Cellular Clock Theory
The cellular clock theory posits that the human lifespan is closely tied to the finite capacity of cells to divide, a phenomenon governed by telomeres, which are protective caps at the ends of...
Cellular Clock Theory
The cellular clock theory posits that the human lifespan is closely tied to the finite capacity of cells to divide, a phenomenon governed by telomeres, which are protective caps at the ends of...
01:37Mitochondria
Mitochondria are eukaryotic cellular organelles that are known to produce energy through a process called oxidative phosphorylation. Besides their primary function, mitochondria are involved in various cellular processes, including cell growth, differentiation, signaling, metabolism, and senescence. Age-related changes cause a decline in mitochondrial quality and integrity due to increased mitochondrial mutations and oxidative damage. Thus, aging can severely impact mitochondrial functions,...