Shared immune biomarkers in necrotizing enterocolitis and neonatal sepsis identified via bioinformatics and machine learning

Necrotizing enterocolitis (NEC) and neonatal sepsis (NS) are major causes of morbidity and mortality in preterm infants, yet their shared molecular basis remains poorly defined. In this study, we integrated two public transcriptomic datasets and applied differential expression analysis, weighted gene co-expression network analysis (WGCNA), and three machine learning algorithms (LASSO, random forest, and XGBoost) to identify shared biomarkers. Four immune-related biomarkers (MAP2K6, CHKA, CA4, and ENTPD7) were identified and used to construct diagnostic models with high performance (AUC = 0.864 for NS; 1.000 for NEC). Immune infiltration analysis revealed distinct immune cell signatures and strong correlations with the selected biomarkers. Regulatory network construction further uncovered potential transcriptional and post-transcriptional regulatory mechanisms. These findings suggest a common immune-related pathogenesis underlying NEC and NS and highlight shared biomarkers with strong diagnostic potential. This integrative analysis provides a foundation for improved early diagnosis and targeted interventions in neonatal care.