Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution
PurposeThe aim of this study was to investigate the activation of immune-metabolic pathways in osteoarthritis (OA) and their role in disease progression. We employed differential expression analysis and Gene Set Enrichment AnalysisMaterials and methodsGene set enrichment analysis (GSEA) to identify...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-06-01
|
| Series: | Frontiers in Immunology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fimmu.2025.1599930/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850100380546891776 |
|---|---|
| author | Hui He Xiumei Zhao Bo Zhang Shijian Zhao Yinteng Wu |
| author_facet | Hui He Xiumei Zhao Bo Zhang Shijian Zhao Yinteng Wu |
| author_sort | Hui He |
| collection | DOAJ |
| description | PurposeThe aim of this study was to investigate the activation of immune-metabolic pathways in osteoarthritis (OA) and their role in disease progression. We employed differential expression analysis and Gene Set Enrichment AnalysisMaterials and methodsGene set enrichment analysis (GSEA) to identify activated immune-metabolism pathways in OA. Subsequently, Weighted gene co-expression network analysis (WGCNA) was used to identify gene modules associated with OA and immune-metabolism scores, followed by enrichment analysis to reveal the functional characteristics of these modules. To identify immune-metabolism related differentially expressed genes (DEGs), we utilized seven machine learning methods, including lasso regression, random forest, bagging, gradient boosting machines (GBM), Xgboost-xgbLinear, Xgboost-xgbtree, and decision trees, to construct predictive models and validate their reliability. Based on the expression profiles of hub immune-metabolism related DEGs, we stratified OA patients into two immune-metabolism related subgroups and deeply investigated the differences in immune profiles, drug responses, functions, and pathways between these subgroups. Additionally, we analyzed the expression and pseudotime trajectories of hub immune-metabolism related DEGs at the single-cell level. Through genome-wide association studies (GWAS), we explored the mechanisms of action of hub immune-metabolism related DEGs. Finally, real-time polymerase chain reaction (RT-PCR) was utilized to verify the expression of hub immune-metabolism related DEGs.ResultsImmune-metabolism related pathways were significantly activated during the development of OA. Thirteen central immune metabolism-related genes (CX3CR1, ADIPOQ, IL17RA, APOD, EGFR, SPP1, PLA2G2A, CXCL14, RARB, ADM, CX3CL1, TNFSF10, and MPO) were identified. Predictive modeling by constructing these genes has good predictive power for identifying OA. These genes are mainly associated with endothelial cells. Single-cell analysis showed that they were all expressed in single cells and varied with cell differentiation. RT-PCR results suggested that they were all significantly expressed in OA.ConclusionOur findings indicate that immune metabolism plays a key role in the development of OA and provide new perspectives for future therapeutic strategies |
| format | Article |
| id | doaj-art-c88e3b33f4cd4b0393cdb6fcffc34102 |
| institution | DOAJ |
| issn | 1664-3224 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Immunology |
| spelling | doaj-art-c88e3b33f4cd4b0393cdb6fcffc341022025-08-20T02:40:18ZengFrontiers Media S.A.Frontiers in Immunology1664-32242025-06-011610.3389/fimmu.2025.15999301599930Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolutionHui He0Xiumei Zhao1Bo Zhang2Shijian Zhao3Yinteng Wu4Department of Orthopedic and Trauma Surgery, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, ChinaSchool of Clinical Medicine, Youjiang Medical University for Nationalities, Baise, Guangxi, ChinaDepartment of Orthopedic and Trauma Surgery, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, ChinaGraduate School, Kunming Medical University, Kunming, Yunnan, ChinaDepartment of Orthopedic and Trauma Surgery, The Third Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, ChinaPurposeThe aim of this study was to investigate the activation of immune-metabolic pathways in osteoarthritis (OA) and their role in disease progression. We employed differential expression analysis and Gene Set Enrichment AnalysisMaterials and methodsGene set enrichment analysis (GSEA) to identify activated immune-metabolism pathways in OA. Subsequently, Weighted gene co-expression network analysis (WGCNA) was used to identify gene modules associated with OA and immune-metabolism scores, followed by enrichment analysis to reveal the functional characteristics of these modules. To identify immune-metabolism related differentially expressed genes (DEGs), we utilized seven machine learning methods, including lasso regression, random forest, bagging, gradient boosting machines (GBM), Xgboost-xgbLinear, Xgboost-xgbtree, and decision trees, to construct predictive models and validate their reliability. Based on the expression profiles of hub immune-metabolism related DEGs, we stratified OA patients into two immune-metabolism related subgroups and deeply investigated the differences in immune profiles, drug responses, functions, and pathways between these subgroups. Additionally, we analyzed the expression and pseudotime trajectories of hub immune-metabolism related DEGs at the single-cell level. Through genome-wide association studies (GWAS), we explored the mechanisms of action of hub immune-metabolism related DEGs. Finally, real-time polymerase chain reaction (RT-PCR) was utilized to verify the expression of hub immune-metabolism related DEGs.ResultsImmune-metabolism related pathways were significantly activated during the development of OA. Thirteen central immune metabolism-related genes (CX3CR1, ADIPOQ, IL17RA, APOD, EGFR, SPP1, PLA2G2A, CXCL14, RARB, ADM, CX3CL1, TNFSF10, and MPO) were identified. Predictive modeling by constructing these genes has good predictive power for identifying OA. These genes are mainly associated with endothelial cells. Single-cell analysis showed that they were all expressed in single cells and varied with cell differentiation. RT-PCR results suggested that they were all significantly expressed in OA.ConclusionOur findings indicate that immune metabolism plays a key role in the development of OA and provide new perspectives for future therapeutic strategieshttps://www.frontiersin.org/articles/10.3389/fimmu.2025.1599930/fullosteoarthritis (OA)immune-metabolismweighted gene co-expression network analysis (WGCNA)machine learninggenome-wide association studies |
| spellingShingle | Hui He Xiumei Zhao Bo Zhang Shijian Zhao Yinteng Wu Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution Frontiers in Immunology osteoarthritis (OA) immune-metabolism weighted gene co-expression network analysis (WGCNA) machine learning genome-wide association studies |
| title | Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution |
| title_full | Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution |
| title_fullStr | Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution |
| title_full_unstemmed | Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution |
| title_short | Integrated multi-omics and machine learning reveals immune-metabolic signatures in osteoarthritis: from bulk RNA-seq to single-cell resolution |
| title_sort | integrated multi omics and machine learning reveals immune metabolic signatures in osteoarthritis from bulk rna seq to single cell resolution |
| topic | osteoarthritis (OA) immune-metabolism weighted gene co-expression network analysis (WGCNA) machine learning genome-wide association studies |
| url | https://www.frontiersin.org/articles/10.3389/fimmu.2025.1599930/full |
| work_keys_str_mv | AT huihe integratedmultiomicsandmachinelearningrevealsimmunemetabolicsignaturesinosteoarthritisfrombulkrnaseqtosinglecellresolution AT xiumeizhao integratedmultiomicsandmachinelearningrevealsimmunemetabolicsignaturesinosteoarthritisfrombulkrnaseqtosinglecellresolution AT bozhang integratedmultiomicsandmachinelearningrevealsimmunemetabolicsignaturesinosteoarthritisfrombulkrnaseqtosinglecellresolution AT shijianzhao integratedmultiomicsandmachinelearningrevealsimmunemetabolicsignaturesinosteoarthritisfrombulkrnaseqtosinglecellresolution AT yintengwu integratedmultiomicsandmachinelearningrevealsimmunemetabolicsignaturesinosteoarthritisfrombulkrnaseqtosinglecellresolution |