Human gut microbiome aging clocks based on taxonomic and functional signatures through multi-view learning

Area of Science:

• Microbiology
• Genomics
• Gerontology

Background:

• The human gut microbiome plays a crucial role in the aging process.
• Current methods for age prediction using gut microbiome metagenomics data are limited.
• Geographical factors can influence gut microbiome composition.

Purpose of the Study:

• To develop a reliable age prediction model using gut microbiome metagenomics data.
• To investigate the influence of geographical factors on gut microbiome-based age prediction.
• To identify microbial biomarkers and metabolic pathways associated with aging.

Main Methods:

• Constructed an age prediction model using 2604 filtered gut microbiome metagenomics datasets.
• Developed an ensemble model with multiple heterogeneous algorithms for multi-view learning.
• Integrated species and pathway profiles, adjusting for host confounding factors.

Main Results:

• The developed model achieved high accuracy in age prediction (R² = 0.599, MAE = 8.33 years).
• Identified specific microbial biomarkers, including *Finegoldia magna*, *Bifidobacterium dentium*, and *Clostridium clostridioforme*, with increased abundance in the elderly.
• Observed significant age-related changes in gut microbiome amino acid utilization, linked to malnutrition and inflammation risks.

Conclusions:

• The study presents a robust method for age prediction leveraging gut microbiome metagenomics.
• Identified key microbial players and metabolic functions indicative of the aging process.
• The findings support the comprehensive utilization of multi-omics data for understanding aging and developing targeted interventions.