Scholarlink

Background <p>Acute myocardial infarction (AMI) is a leading cause of death worldwide, with oxidative stress (OS) playing a central role in its pathogenesis. However, the dynamic expression profiles and functional implications of OS-related genes at the single-cell level remain poorly understood.</p> Methods <p>OS activity first assessed in single-cell RNA sequencing (scRNA-seq) data from myocardial tissue samples. Then, we integrated scRNA-seq data from AMI patients with multi-cohort bulk RNA-seq datasets to comprehensively assess OS activity across peripheral immune cells. A curated list of 807 OS-related genes was used to construct OS activity scores at the single-cell level. Differential analysis, correlation filtering, and univariate logistic regression were applied to identify high-risk OS-related genes. Six machine learning algorithms-including LASSO, Random Forest, Boruta, Bayesian modeling, LVQ, and Treebag were used in consensus to prioritize diagnostic biomarkers. To further characterize the functional implications of TKT, we performed downstream analyses including Western Blot analysis, single-cell expression profiling, pseudotime trajectory inference, and cell–cell communication network construction.</p> Results <p>TKT emerged as a robust diagnostic marker, consistently selected across all machine learning models and validated in both training and independent cohorts (AUC &gt; 0.73). TKT was predominantly expressed in monocytes and neutrophils theoretically, where it correlated with high OS activity and pro-inflammatory gene signatures. Pseudotime analysis revealed distinct expression dynamics in monocyte and neutrophil lineages. CellChat analysis further suggested that TKT⁺ immune cells act as central hubs within the AMI immune microenvironment, mediating enhanced inflammatory communication.</p> Conclusion <p>This study identifies TKT as a high-confidence, oxidative stress-associated biomarker for AMI through an integrated multi-omics and machine learning approach. TKT may serve as a bridge between metabolic reprogramming and immune activation, offering insights into potential therapeutic targets in post-infarction inflammation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Single-cell and machine learning-integrated bulk RNA-seq analysis reveals TKT as an oxidative stress-associated diagnostic biomarker in acute myocardial infarction

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月7日

查看全文
  • Jiayi Wang,
  • Shouwen Zhang,
  • Fahong Li,
  • Yanjin Wei,
  • Ming Li

摘要

Background

Acute myocardial infarction (AMI) is a leading cause of death worldwide, with oxidative stress (OS) playing a central role in its pathogenesis. However, the dynamic expression profiles and functional implications of OS-related genes at the single-cell level remain poorly understood.

Methods

OS activity first assessed in single-cell RNA sequencing (scRNA-seq) data from myocardial tissue samples. Then, we integrated scRNA-seq data from AMI patients with multi-cohort bulk RNA-seq datasets to comprehensively assess OS activity across peripheral immune cells. A curated list of 807 OS-related genes was used to construct OS activity scores at the single-cell level. Differential analysis, correlation filtering, and univariate logistic regression were applied to identify high-risk OS-related genes. Six machine learning algorithms-including LASSO, Random Forest, Boruta, Bayesian modeling, LVQ, and Treebag were used in consensus to prioritize diagnostic biomarkers. To further characterize the functional implications of TKT, we performed downstream analyses including Western Blot analysis, single-cell expression profiling, pseudotime trajectory inference, and cell–cell communication network construction.

Results

TKT emerged as a robust diagnostic marker, consistently selected across all machine learning models and validated in both training and independent cohorts (AUC > 0.73). TKT was predominantly expressed in monocytes and neutrophils theoretically, where it correlated with high OS activity and pro-inflammatory gene signatures. Pseudotime analysis revealed distinct expression dynamics in monocyte and neutrophil lineages. CellChat analysis further suggested that TKT⁺ immune cells act as central hubs within the AMI immune microenvironment, mediating enhanced inflammatory communication.

Conclusion

This study identifies TKT as a high-confidence, oxidative stress-associated biomarker for AMI through an integrated multi-omics and machine learning approach. TKT may serve as a bridge between metabolic reprogramming and immune activation, offering insights into potential therapeutic targets in post-infarction inflammation.