Identifying semaphorin 3C as a biomarker for sarcopenia and coronary artery disease via bioinformatics and machine learning.

Objective: Sarcopenia not only affects patients' quality of life but also may exacerbate the pathological processes of coronary artery disease (CAD). This study aimed to identify potential biomarkers to improve the combined diagnosis and treatment of sarcopenia and CAD.

Methods: Datasets for sarcopenia and CAD were sourced from the Gene Expression Omnibus (GEO). Weighted gene co-expression network analysis (WGCNA) was used to identify key module genes. Functional enrichment analysis was conducted to explore biological significance. Three machine learning algorithms were applied to further determine candidate hub genes, including SVM-RFE, LASSO regression, and random forest (RF). Then, we generated receiver operating characteristic (ROC) curves to evaluate the diagnostic efficacy of the candidate genes. Moreover, mendelian randomization (MR) analysis was conducted based on GWAS summary data, along with sensitivity analysis to explore causal relationships.

Results: WGCNA analysis identified 278 genes associated with sarcopenia and CAD. The results of the enrichment analysis indicated a complex interplay between RNA metabolism, signaling pathways, and cellular stress responses. Through machine learning methods and ROC curves, we identified the key gene semaphorin 3C (SEMA3C). MR analysis revealed that higher plasma levels of SEMA3C are associated with an increased risk of CAD (OR = 1.068, 95 % CI 1.012-1.128, P = 0.016) and low hand grip strength (HGS) (OR = 1.059, 95 % CI 1.010-1.110, P = 0.018) .

Conclusion: SEMA3C has been identified as a key gene for sarcopenia and CAD. This insight suggests that targeting SEMA3C may offer new therapeutic opportunities in related conditions.