Identification of arachidonic acid metabolism-related diagnostic markers in 1 heart failure based on bioinformatics analysis and machine learning

Saiqing ChenChunxia ZhangYueting Yu^*

• Yiwu Hospital of Traditional Chinese Medicine, Yiwu, China

Background 12 Heart failure (HF) represents the terminal phase of multiple cardiovascular conditions 13 and is associated with significant morbidity and mortality rates. Arachidonic acid 14 (AA), an essential fatty acid, plays a crucial role in modulating cardiovascular 15 function under both normal and disease states. The purpose of this research was to 16 examine how AA is related to HF, providing new perspective for individualized 17 treatment. 18 Methods 19 Transcriptomic datasets were retrieved from the Gene Expression Omnibus (GEO) 20 database. The raw data were consolidated to identify differentially expressed genes 21 (DEGs) and subsequently subjected to bioinformatics analysis. Gene ontology (GO) 22 annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment 23 analyses were performed. Signature genes were identified through Least Absolute 24 Shrinkage and Selection Operator (LASSO) regression, Support Vector 25 Machine-Recursive Feature Elimination (SVM-RFE), and Random Forest (RF) 26 algorithms. Receiver Operating Characteristic (ROC) curves were generated for gene 27 evaluation, and a nomogram was developed. An analysis of immune cell infiltration 28 was conducted using Single Sample Gene Set Enrichment Analysis (ssGSEA), and 29 Gene Set Enrichment Analysis (GSEA) was conducted to determine important 30 pathways. Subsequently, we also performed drug sensitivity evaluation. Finally, the 31 expression levels of the identified signature genes in HF samples were confirmed 32 using qRT-PCR analysis. 33 Results 34 Four characteristic genes demonstrating favorable performance in the ROC analysis. 35 The comprehensive nomogram developed in this study exhibited enhanced clinical 36 utility. In addition, notable variations in immune cell infiltration levels were detected, 37 and GSEA highlighted key biological pathways. 38 Conclusion 39 This investigation demonstrated a strong association between arachidonic 40 acid-associated gene expression and heightened risk of HF, offering novel 41 perspectives on the disease's underlying pathological processes and providing 42 potential insights for personalized management of HF.