Biomarkers identified through integrated bioinformatics approaches often exhibit limited temporal stability during sepsis progression. To address this limitation, temporal changes in the serum transcriptome of sepsis patients were analyzed at multiple time points to investigate underlying pathological mechanisms. The datasets GSE54514 and GSE212092 from the Gene Expression Omnibus (GEO) database were used for gene identification and validation. The key marker RAPGEF6 with temporal characteristics was identified by integrating time-series clustering analysis, differential expression analysis, weighted gene co-expression network analysis (WGCNA), and single-cell data analysis. Stable RAPGEF6-knockdown RAW264.7 macrophages were generated using lentiviral shRNA delivery to investigate the role of RAPGEF6 in sepsis. A total of 2,701 genes showed time-associated expression patterns during sepsis progression. Integrative analysis combining differential expression and WGCNA-derived modules identified three hub genes: RAPGEF6, CHD9, and POGULT1. RAPGEF6 showed the highest area under the receiver operating characteristic curve (AUROC) of 0.9206 (95% CI: 0.8383-1). Immune infiltration analysis indicated that RAPGEF6 had the strongest correlation with M1 macrophages (correlation coefficient = 0.42). The analysis of single-cell data targeting key genes and external experiments confirmed this finding. RAPGEF6 expression in macrophages increased during the early phase of lipopolysaccharide (LPS) stimulation and then decreased over time. Stable knockout of RAPGEF6 in macrophages using lentiviral vectors attenuated LPS-induced inflammatory responses and oxidative stress by inhibiting NF-κB-mediated necroptosis. Time-series transcriptomic analysis revealed highly dynamic gene expression trajectories in sepsis patients. RAPGEF6, identified as a key biomarker with temporal characteristics, may regulate inflammation and oxidative stress via NF-κB-mediated necroptosis.