Machine learning-mediated identification of ferroptosis-related genes in osteonecrosis of the femoral head

Abstract

Osteonecrosis of the femoral head (ONFH) is a condition caused by a disruption or damage to the femoral head's blood supply, which causes the death of bone cells and bone marrow components and prevents future regeneration. Ferroptosis, a type of controlled cell death, is caused by iron-dependent lipid peroxidation. Here, we identified ferroptosis-related genes and infiltrating immune cells involved in ONFH and predicted the underlying molecular mechanisms. The GSE123568 dataset was subjected to differential expression analysis to identify genes related to ferroptosis. Subsequently, GO and KEGG pathway enrichment analyses, as well as protein-protein interaction (PPI) network analysis, were conducted. Hub genes involved in ferroptosis were identified using machine learning and other techniques. Additionally, immune infiltration analysis and lncRNA-miRNA-mRNA network prediction analysis were performed. Finally, we determined whether ferroptosis occurred by measuring iron content. The hub genes were validated by ROC curve analysis and qRT-PCR. Four ferroptosis-related hub genes (MAPK3, PTGS2, STK11, and SLC2A1) were identified. Additionally, immune infiltration analysis revealed a strong correlation among ONFH, hub genes, and various immune cells. Finally, we predicted the network relationship between differentially expressed lncRNAs and hub genes in the lncRNA-miRNA-mRNA network. MAPK3, PTGS2, STK11, and SLC2A1 have been identified as potential ferroptosis-related biomarkers and drug targets for the diagnosis and prognosis of ONFH, while some immune cells, as well as the interaction between lncRNA, miRNA, and mRNA, have also been identified as potential pathogenesis markers and therapeutic targets.

Keywords: bioinformatics analysis; ferroptosis; immune infiltration; machine learning; osteonecrosis of the femoral head.