Identification of biomarkers and immune microenvironment associated with heart failure through bioinformatics and machine learning.

Background: Heart failure (HF) is the end stage of various cardiovascular diseases. Identifying new biomarkers is essential for early diagnosis, prognosis, and treatment. This study applied bioinformatics to identify potential HF biomarkers and explore the role of the immune microenvironment. Methods: Gene expression data were obtained from the Gene Expression Omnibus (GEO) database. Differential expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA) were used to identify key genes. Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis were performed. Feature genes were further determined using two machine learning algorithms, Random Forest (RF) and Least Absolute Shrinkage and Selection Operator (LASSO), with diagnostic accuracy assessed via Receiver Operating Characteristic (ROC) curves and nomograms to screen hub genes, and external datasets further were used for validation. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) was used to validate the expression levels of hub genes in clinical samples. Single Sample Gene Set Enrichment Analysis and CIBERSORT algorithm were applied to evaluate immune cell infiltration in HF and its relationship with hub genes. Results: Differential analysis identified 165 differentially expressed genes (DEGs), and WGCNA revealed the "blue" module showing a significant correlation with HF. Integration of the DEGs and the "blue" module genes identified 28 common genes. KEGG pathway enrichment analysis suggested that these genes may be involved in the cytoskeleton in muscle cells pathway. Lasso and RF algorithms confirmed 7 key genes as potential biomarkers for HF, and further analysis using the ROC curve identified 4 hub genes with good diagnostic value, namely, High mobility group N 2 ( Conclusion: This study identifies