Single-cell sequencing analysis and multiple machine learning methods identified immune-associated SERPINB1 and CPEB4...

Acute respiratory distress syndrome (ARDS) is a life-threatening complication of COVID-19, often resulting in respiratory failure and high mortality. Identifying effective molecular biomarkers is crucial for understanding its pathogenesis and improving diagnosis and treatment strategies. We analyzed transcriptomic and single-cell RNA-seq data from public datasets (GSE172114, GSE149878, and GSE213313). Differentially expressed genes (DEGs) were identified using the limma package and weighted gene co-expression network analysis (WGCNA). Single-cell analysis was used to define cell-type-specific expression. Three machine learning algorithms-LASSO, SVM-RFE, and Random Forest-were applied to identify robust hub genes. External dataset GSE213313 was used for validation. CIBERSORT was applied to estimate immune cell infiltration in ARDS tissues. We identified 915 DEGs between COVID-19-induced ARDS and controls, mainly enriched in immune receptor activity and cytokine signaling. Through integrative machine learning and validation, SERPINB1 and CPEB4 were identified as key genes, with strong diagnostic performance (AUCs: 0.940 and 0.948, respectively). Immune infiltration analysis revealed that both genes were highly correlated with neutrophils, and also associated with B memory cells, T cells, NK cells, monocytes, and mast cells. GSEA showed these genes were involved in immune and inflammatory pathways, indicating functional relevance in ARDS. SERPINB1 and CPEB4 were identified as novel immune-related biomarkers for COVID-19-induced ARDS. Their strong association with neutrophil infiltration suggests that they may play critical roles in disease progression. These findings provide new insights into immune mechanisms and offer promising targets for early diagnosis and therapeutic intervention in ARDS.