Department of Geriatric Medicine, Qilu Hospital, Cheeloo College of Medicine, Shandong University, Jinan, China.
Cancer Institute, Qingdao University, Qingdao, 266071, China.
Sci Rep. 2024 Jan 23;14(1):2026. doi: 10.1038/s41598-024-51536-3.
Sepsis is a major global health problem, causing a significant burden of disease and death worldwide. Risk stratification of sepsis patients, identification of severe patients and timely initiation of treatment can effectively improve the prognosis of sepsis patients. We procured gene expression datasets for sepsis (GSE54514, GSE65682, GSE95233) from the Gene Expression Omnibus and performed normalization to mitigate batch effects. Subsequently, we applied weighted gene co-expression network analysis to categorize genes into modules that exhibit correlation with macrophage activity. To pinpoint macrophage-associated genes (MAAGs), we executed differential expression analysis and single sample gene set enrichment analysis. We then established a prognostic model derived from four MAAGs that were significantly differentially expressed. Functional enrichment analysis and immune infiltration assessments were instrumental in deciphering the biological mechanisms involved. Furthermore, we employed principal component analysis and conducted survival outcome analyses to delineate molecular subgroups within sepsis. Four novel MAAGs-CD160, CX3CR1, DENND2D, and FAM43A-were validated and used to create a prognostic model. Subgroup classification revealed distinct molecular profiles and a correlation with 28-day survival outcomes. The MAAGs risk score was developed through univariate Cox, LASSO, and multivariate Cox analyses to predict patient prognosis. Validation of the risk score upheld its prognostic significance. Functional enrichment implicated ribonucleoprotein complex biogenesis, mitochondrial matrix, and transcription coregulator activity in sepsis, with an immune infiltration analysis indicating an association between MAAGs risk score and immune cell populations. The four MAAGs exhibited strong diagnostic capabilities for sepsis. The research successfully developed a MAAG-based prognostic model for sepsis, demonstrating that such genes can significantly stratify risk and reflect immune status. Although in-depth mechanistic studies are needed, these findings propose novel targets for therapy and provide a foundation for future precise clinical sepsis management.
脓毒症是一个全球性的重大健康问题，在全球范围内造成了巨大的疾病和死亡负担。对脓毒症患者进行风险分层、识别重症患者并及时开始治疗，可以有效地改善脓毒症患者的预后。我们从基因表达综合数据库中获取了脓毒症的基因表达数据集（GSE54514、GSE65682、GSE95233），并进行了标准化处理以减轻批次效应。随后，我们应用加权基因共表达网络分析将基因分类为与巨噬细胞活性相关的模块。为了确定与巨噬细胞相关的基因（MAAGs），我们进行了差异表达分析和单样本基因集富集分析。然后，我们建立了一个由四个差异表达的 MAAGs 构建的预后模型。功能富集分析和免疫浸润评估有助于解析涉及的生物学机制。此外，我们还进行了主成分分析和生存结果分析，以描绘脓毒症中的分子亚群。四个新的 MAAGs-CD160、CX3CR1、DENND2D 和 FAM43A-经过验证并用于构建预后模型。亚组分类揭示了不同的分子特征与 28 天生存结局的相关性。MAAGs 风险评分通过单变量 Cox、LASSO 和多变量 Cox 分析来预测患者的预后。风险评分的验证证实了其预后意义。功能富集分析表明，核糖核蛋白复合物生物发生、线粒体基质和转录共激活因子活性与脓毒症有关，免疫浸润分析表明 MAAGs 风险评分与免疫细胞群之间存在相关性。这四个 MAAGs 对脓毒症具有很强的诊断能力。该研究成功开发了基于 MAAG 的脓毒症预后模型，表明这些基因可以显著分层风险并反映免疫状态。虽然需要进行深入的机制研究，但这些发现为治疗提供了新的靶点，并为未来精确的临床脓毒症管理奠定了基础。
