Suppr超能文献

通过机器学习和实验验证识别用于慢性阻塞性肺疾病诊断的新型乳酸化相关生物标志物

Identification of Novel Lactylation-Related Biomarkers for COPD Diagnosis Through Machine Learning and Experimental Validation.

作者信息

Hu Chundi, Qian Weiliang, Wei Runling, Liu Gengluan, Jiang Qin, Sun Zhenglong, Li Hui

机构信息

Department of Respiratory and Critical Care Medicine, The Seventh Affiliated Hospital, Sun Yat-sen University, Shenzhen 518107, China.

Shenzhen Bay Laboratory, Shenzhen 518000, China.

出版信息

Biomedicines. 2025 Aug 18;13(8):2006. doi: 10.3390/biomedicines13082006.

Abstract

This study aims to identify clinically relevant lactylation-related biomarkers in chronic obstructive pulmonary disease (COPD) and investigate their potential mechanistic roles in COPD pathogenesis. Differentially expressed genes (DEGs) were identified from the GSE21359 dataset, followed by weighted gene co-expression network analysis (WGCNA) to detect COPD-associated modules. Least absolute shrinkage and selection operator (LASSO) regression and support vector machine-recursive feature elimination (SVM-RFE) algorithms were applied to screen lactylation-related biomarkers, with diagnostic performance evaluated through the ROC curve. Candidates were validated in the GSE76925 dataset for expression and diagnostic robustness. Immune cell infiltration patterns were exhibited using EPIC deconvolution. Single-cell transcriptomics (from GSE173896) were processed via the 'Seurat' package encompassing quality control, dimensionality reduction, and cell type annotation. Cell-type-specific markers and intercellular communication networks were delineated using the 'FindAllMarkers' package and the 'CellChat' R package, respectively. In vitro validation was conducted using a cigarette smoke extract (CSE)-induced COPD model. Integrated transcriptomic approaches and multi-algorithm screening (LASSO/Boruta/SVM-RFE) revealed carbonyl reductase 1 (CBR1) and peroxiredoxin 1 (PRDX1) as core COPD biomarkers enriched in oxidation-reduction and inflammatory pathways, with high diagnostic accuracy (AUC > 0.85). Immune profiling and scRNA-seq delineated macrophage and cancer-associated fibroblasts (CAFs) infiltration with oxidative-redox transcriptional dominance in COPD. CBR1 was significantly upregulated in T cells, neutrophils, and mast cells; and PRDX1 showed significant upregulation in endothelial, macrophage, and ciliated cells. Experimental validation in CSE-induced models confirmed significant upregulation of both biomarkers via transcription PCR (qRT-PCR) and immunofluorescence. CBR1 and PRDX1 are lactylation-associated diagnostic markers, with lactylation-driven redox imbalance implicated in COPD progression.

摘要

本研究旨在鉴定慢性阻塞性肺疾病(COPD)中与乳酰化相关的临床相关生物标志物,并研究它们在COPD发病机制中的潜在作用机制。从GSE21359数据集中鉴定差异表达基因(DEGs),随后进行加权基因共表达网络分析(WGCNA)以检测与COPD相关的模块。应用最小绝对收缩和选择算子(LASSO)回归和支持向量机递归特征消除(SVM-RFE)算法筛选与乳酰化相关的生物标志物,并通过ROC曲线评估诊断性能。在GSE76925数据集中对候选物进行表达和诊断稳健性验证。使用EPIC反卷积展示免疫细胞浸润模式。单细胞转录组学(来自GSE173896)通过“Seurat”软件包进行处理,包括质量控制、降维和细胞类型注释。分别使用“FindAllMarkers”软件包和“CellChat”R软件包描绘细胞类型特异性标记物和细胞间通讯网络。使用香烟烟雾提取物(CSE)诱导的COPD模型进行体外验证。综合转录组学方法和多算法筛选(LASSO/Boruta/SVM-RFE)揭示羰基还原酶1(CBR1)和过氧化物酶1(PRDX1)作为核心COPD生物标志物,富集于氧化还原和炎症途径,具有较高的诊断准确性(AUC>0.85)。免疫分析和单细胞RNA测序描绘了COPD中巨噬细胞和癌症相关成纤维细胞(CAFs)的浸润以及氧化还原转录优势。CBR1在T细胞、中性粒细胞和肥大细胞中显著上调;PRDX1在内皮细胞、巨噬细胞和纤毛细胞中显著上调。在CSE诱导的模型中的实验验证通过转录PCR(qRT-PCR)和免疫荧光证实了两种生物标志物的显著上调。CBR1和PRDX1是与乳酰化相关的诊断标志物,乳酰化驱动的氧化还原失衡与COPD进展有关。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/055d/12383316/36d7f2fa1b31/biomedicines-13-02006-g001.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验