BACKGROUND Acute lung injury (ALI) after liver transplantation (LT) is a critical complication negatively affecting transplant outcomes and patient survival. However, effective biomarkers for early diagnosis remain unidentified. This study aimed to identify hub biomarkers and signaling pathways involved in post-LT ALI through integrated bioinformatics and machine-learning analyses and to validate their diagnostic potential. MATERIAL AND METHODS Differential gene expression analysis identified 27 differentially expressed genes (DEGs). Functional enrichment analyses revealed significant involvement in cytokine-mediated signaling, particularly within the NF-kB and TNF pathways. Single-sample gene set enrichment analysis (ssGSEA) evaluated immune infiltration. Machine-learning algorithms identified crucial biomarkers for ALI prediction. Transcription factor-hub gene and competitive endogenous RNA (ceRNA) networks were constructed. Single-cell RNA sequencing (scRNA-seq) and enzyme-linked immunosorbent assay (ELISA) analyses validated biomarker expression patterns in relation to ALI. RESULTS Hub biomarkers identified included CXCL3, CD48, and IRAK3. ssGSEA revealed prominent macrophage and neutrophil infiltration associated with ALI. Machine-learning models confirmed CXCL3, CD48, and IRAK3 as reliable predictive biomarkers, which were incorporated into a robust diagnostic nomogram. scRNA-seq analysis showed cell-type-specific expression patterns for CXCL3, CD48, and IRAK3, with heterogeneous associations across datasets. ELISA validated significantly altered protein levels of CXCL3, CD48, and IRAK3 in post-transplant ALI cases compared with controls. CONCLUSIONS CXCL3, CD48, and IRAK3 are novel and promising diagnostic biomarkers for predicting post-LT ALI. These findings provide foundational insights that could support improved diagnosis, prevention strategies, and targeted therapeutic interventions, ultimately enhancing patient outcomes after liver transplantation.
Guiting et al. (Mon,) studied this question.