Identification and Analysis of Exercise-Related Gene Biomarkers in Sarcopenia

Objective: To explore the potential exercise-associated biomarkers in sarcopenia, and reveal the possible molecular mechanism of the exercise improved sarcopenia. Methods: The Gene Expression Omnibus database was utilized to get datasets GSE28422 and GSE125830. GSE28422 served as the training set for biomarker screening, while GSE125830 functioned as the validation set. Common genes were identified using differential expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA), followed by an enrichment analysis of the overlapping genes. Subsequently, machine learning methods were used to select the best feature genes, and the predictive performance was evaluated using the ROC curve to identify the potential gene biomarkers. Based on these findings, the regulatory network of gene biomarkers was explored, and the related immune characteristics were analyzed. Results: Differential expression analysis and WGCNA yielded 92 common genes, which a functional enrichment study linked to the processes of muscle and organ development, as well as immune and inflammatory response pathways. Machine learning further screening and validation set validation identified potential gene markers such as FNIP2, NFKBIA, TNFRSF12A and NR4A3. Additionally, Immune cell infiltration analysis revealed that the identified gene biomarkers all have a strong correlation with eosinophils. Conclusion: FNIP2, NFKBIA, TNFRSF12A, and NR4A3 could be essential biomarkers in related to the development and prognosis of sarcopenia.