Related Experiment Video
Updated: Sep 16, 2025
08:51
Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
1.5K
Machine learning combined multi-omics analysis to explore key oxidative stress features in systemic lupus
Hongwei Zhou1, Xiaoqing Li1, Yanyu Zhang1
1Department of Laboratory Diagnosis, The First Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China.
Frontiers in Immunology
|July 8, 2025
Summary
Systemic lupus erythematosus (SLE) involves metabolic dysregulation and oxidative stress (OS). This study identified six key genes linked to OS and immune cells, offering potential for early SLE detection and targeted therapies.
Area of Science:
- Immunology
- Metabolomics
- Systems Biology
- Bioinformatics
Background:
- Metabolic dysregulation and redox imbalance in immune cells are central to systemic lupus erythematosus (SLE) pathogenesis.
- Understanding the interplay between oxidative stress (OS) and metabolic pathways in SLE is crucial for developing effective treatments.
Purpose of the Study:
- To explore critical oxidative stress (OS) features and their interrelationships in systemic lupus erythematosus (SLE) pathogenesis.
- To identify key genes and metabolic pathways associated with OS in SLE using multi-omics and machine learning approaches.
Main Methods:
- Analysis of transcriptomic datasets using Gene Set Variation Analysis (GSVA) to identify SLE- and OS-associated pathways.
- Application of multiple machine learning algorithms (DL, RF, XGBoost, SVM, LASSO) for OS-related gene prediction.
- Assessment of immune infiltration, single-cell transcriptomic patterns, and metabolomics profiling (LC-MS) in SLE patients and healthy controls.
Main Results:
- Fifteen metabolic pathways were significantly linked to SLE, with seven strongly associated with OS and energy metabolism.
- Machine learning identified six key OS-related genes (ABCB1, AKR1C3, EIF2AK2, IFIH1, NPC1, SCO2) with robust predictive performance and correlations with immune cell subsets.
- SLE patients exhibited higher OS levels, lower antioxidant stress (AOS) levels, and distinct gene expression patterns in peripheral blood mononuclear cells (PBMCs).
Conclusions:
- This multi-omics and machine learning investigation revealed significant disruptions in OS-related metabolic pathways and metabolites in SLE.
- The identified six key genes demonstrate diagnostic and therapeutic potential for SLE, highlighting their association with OS/AOS levels and metabolites.
- Findings provide a foundation for early detection and targeted treatment strategies for systemic lupus erythematosus.
Keywords:
machine learningmetabolomicsoxidative stresssingle-cell transcriptomicsystemic lupus erythematosustranscriptomics
