Machine Learning-Based Identification of Molecular Signatures in PTOA Cell Subtypes via Single-Cell Transcriptomics in a Mouse Model
Stem Cell Reviews and Reports · 2026-07 · 22 (5) · 2419-2436
DOI: 10.1007/s12015-026-11121-9 ↗Kurz gefasst
Fachartikel in der Zeitschrift Stem Cell Reviews and Reports, online erschienen am 14.04.2026. Verfasst von Dujiang Yang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Chengdu University of Traditional Chinese medicine, No. 37, Shierqiao Road, Chengdu, 610075, Sichuan Province, P.R. China. Hauptthemen laut MEDLINE: Machine Learning, Osteoarthritis, Transcriptome, Single-Cell Analysis. Die Arbeit nennt 50 Literaturangaben.
Abstract (Originalsprache)
OBJECTIVES: Osteoarthritis (OA) is the most prevalent joint disorder, whereas post-traumatic osteoarthritis (PTOA) denotes a form of arthritis that arises secondary to acute joint injury. METHODS: This was a combined bioinformatics and experimental validation study. We first analyzed single-cell RNA sequencing data (GSE200843) from murine knee joints following anterior cruciate ligament rupture to map granulocyte subpopulations. High-dimensional weighted gene co-expression network analysis (hdWGCNA) was used to identify PTOA-associated gene modules. Three machine learning algorithms (LASSO, SVM-RFE, and Random Forest) were applied to screen hub genes, followed by validation in external datasets (GSE26475, GSE112641) and by qRT-PCR in a mouse model (n = 6 per group). RESULTS: We identified five distinct granulocyte subpopulations, one of which (OA granulocytes) was significantly expanded in PTOA tissues. Through intersection of hdWGCNA-derived module genes and differentially expressed genes, combined with machine learning, Tgfbi and Mpp7 were identified as hub genes. These biomarkers could be developed into diagnostic assays for patients at risk of PTOA following acute joint injury. qRT-PCR confirmed that Tgfbi was significantly upregulated (p = 0.0014) and Mpp7 downregulated (p < 0.0002) in synovial tissues of the PTOA mouse model compared to controls. Immune infiltration analysis revealed significant correlations of these hub genes with naive B cells and M1 macrophages. CONCLUSIONS: This study identifies Tgfbi and Mpp7 as potential diagnostic biomarkers for PTOA. The findings suggest that assessing the expression of these genes may aid in early diagnosis and risk stratification, potentially enabling timely therapeutic intervention before irreversible joint damage occurs. These biomarkers could be developed into diagnostic assays for patients at risk of PTOA following acute joint injury.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Chengdu University of Traditional Chinese medicine, No. 37, Shierqiao Road, Chengdu, 610075, Sichuan Province, P.R. China.
- Affiliated Hospital of Chengdu University of Traditional Chinese medicine, No. 39, Shierqiao Road, Jinniu District, Chengdu, 610075, Sichuan Province, P.R. China.
- Department of Joint Surgery, The Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University, NO.182, Chunhui Road, Longmatan District, Luzhou, 646000, Sichuan, China.
- Department of Joint Surgery, The Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University, NO.182, Chunhui Road, Longmatan District, Luzhou, 646000, Sichuan, China. wang_guoyou1981@163.com.
- Center for Orthopedic Diseases Research, The Affiliated Traditional Chinese Medicine Hospital, Southwest Medical University, Luzhou, Sichuan, China. wang_guoyou1981@163.com.
Zitieren
APA
Yang, D., Gong, G., Chen, J., Song, J., Ye, Z., Wang, S., & Wang, G. (2026). Machine Learning-Based Identification of Molecular Signatures in PTOA Cell Subtypes via Single-Cell Transcriptomics in a Mouse Model. Stem Cell Reviews and Reports, 22(5), 2419-2436. https://doi.org/10.1007/s12015-026-11121-9
Vancouver
Yang D, Gong G, Chen J, Song J, Ye Z, Wang S, et al. Machine Learning-Based Identification of Molecular Signatures in PTOA Cell Subtypes via Single-Cell Transcriptomics in a Mouse Model. Stem Cell Reviews and Reports. 2026;22(5):2419-2436. doi:10.1007/s12015-026-11121-9
BibTeX
@article{Yang2026,
author = {Yang, Dujiang and Gong, Gaowen and Chen, Junjie and Song, Jiafeng and Ye, Zhijun and Wang, Shuang and Wang, Guoyou},
title = {Machine Learning-Based Identification of Molecular Signatures in PTOA Cell Subtypes via Single-Cell Transcriptomics in a Mouse Model},
journal = {Stem Cell Reviews and Reports},
year = {2026},
volume = {22},
number = {5},
pages = {2419--2436},
doi = {10.1007/s12015-026-11121-9}
}
EUVIMED