Single-Cell Dissection of Sarcopenia and Aging in Human Skeletal Muscle Reveals Heterogeneous Mechanisms via Deep Transfer Learning

Background: Sarcopenia is characterized by progressive loss of skeletal muscle mass and function, yet conventional bulk RNA sequencing approaches cannot resolve the cellular heterogeneity underlying the transition from physiological aging to pathological muscle decline. Methods: This study employed Diagnostic Evidence GAuge of Single cells (DEGAS), a deep transfer learning framework, to integrate bulk and single-cell RNA-seq data and transfer donor-level clinical information to individual skeletal muscle cells. Sarcopenia- and age-associated scores were assigned to individual cells, followed by differential expression, pathway enrichment, cis-Mendelian randomization (cis-MR), and in silico perturbation analyses. Results: DEGAS revealed marked transcriptional heterogeneity among skeletal muscle cells. Compared with bulk RNA-seq, which identified only four differentially expressed genes, DEGAS identified 134 genes between high- and low-sarcopenia-score cells, showing enhanced cytoplasmic translation (fold enrichment [FE] = 29.85, p < 0.001) accompanied by impaired glycolytic activity (normalized enrichment score [NES] = −2.00, p = 0.02) and muscle adaptation programs. Within high-sarcopenia-score cells, aging was associated with 484 differentially expressed genes, characterized by increased translational programs (FE = 31.97, p < 0.001) and reduced cellular respiration and microenvironment-related functions. Unfavorable age–sarcopenia decoupling identified 95 genes associated with impaired skeletal muscle adaptation (FE = 58.33, p < 0.001). Cis-MR identified eight genetically supported core genes, and in silico perturbation prioritized RPS2, whose knockout altered ribosomal, muscle structural, and energy metabolism-related pathways, including oxidative phosphorylation (NES = 2.18, p < 0.001). Conclusions: This study demonstrates that sarcopenia comprises heterogeneous cellular states driven by distinct mechanisms, including metabolic failure, microenvironmental deterioration, and impaired adaptive muscle function. Integrating genetic evidence with perturbation analysis identifies RPS2 as a potential regulatory node in sarcopenia-associated muscle states.

Authors

Institutions

Publication Details

Journal
Biomedicines
Published
2026-11-16
DOI
https://doi.org/10.3390/biomedicines14112650
Primary Topic
Nutrition and Health in Aging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Single-Cell Dissection of Sarcopenia and Aging in Human Skeletal Muscle Reveals Heterogeneous Mechanisms via Deep Transfer Learning

Peng Ziyi, Bihong Sun, Li Liu, Tian Yu et al.
5 links
Biomedicines
Nutrition and Health in Aging
article

Single-Cell Dissection of Sarcopenia and Aging in Human Skeletal Muscle Reveals Heterogeneous Mechanisms via Deep Transfer Learning

Peng Ziyi, Bihong Sun, Li Liu, Tian Yu, Zhelong Liu, Xinran Liu
article en

Abstract

Background: Sarcopenia is characterized by progressive loss of skeletal muscle mass and function, yet conventional bulk RNA sequencing approaches cannot resolve the cellular heterogeneity underlying the transition from physiological aging to pathological muscle decline. Methods: This study employed Diagnostic Evidence GAuge of Single cells (DEGAS), a deep transfer learning framework, to integrate bulk and single-cell RNA-seq data and transfer donor-level clinical information to individual skeletal muscle cells. Sarcopenia- and age-associated scores were assigned to individual cells, followed by differential expression, pathway enrichment, cis-Mendelian randomization (cis-MR), and in silico perturbation analyses. Results: DEGAS revealed marked transcriptional heterogeneity among skeletal muscle cells. Compared with bulk RNA-seq, which identified only four differentially expressed genes, DEGAS identified 134 genes between high- and low-sarcopenia-score cells, showing enhanced cytoplasmic translation (fold enrichment [FE] = 29.85, p < 0.001) accompanied by impaired glycolytic activity (normalized enrichment score [NES] = −2.00, p = 0.02) and muscle adaptation programs. Within high-sarcopenia-score cells, aging was associated with 484 differentially expressed genes, characterized by increased translational programs (FE = 31.97, p < 0.001) and reduced cellular respiration and microenvironment-related functions. Unfavorable age–sarcopenia decoupling identified 95 genes associated with impaired skeletal muscle adaptation (FE = 58.33, p < 0.001). Cis-MR identified eight genetically supported core genes, and in silico perturbation prioritized RPS2, whose knockout altered ribosomal, muscle structural, and energy metabolism-related pathways, including oxidative phosphorylation (NES = 2.18, p < 0.001). Conclusions: This study demonstrates that sarcopenia comprises heterogeneous cellular states driven by distinct mechanisms, including metabolic failure, microenvironmental deterioration, and impaired adaptive muscle function. Integrating genetic evidence with perturbation analysis identifies RPS2 as a potential regulatory node in sarcopenia-associated muscle states.

BiomedicinesVol. 14(11)
Tongji Hospital (CN), Huazhong University of Science and Technology (CN)
Good health and well-being
Nutrition and Health in Aging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon