Identifying and Diagnosing Lytic Cell Death Genes in Atherosclerosis Using Machine Learning and Bioinformatics
Guolin Zhang,1,* Ruicong Ma,2,* Hongjin Jin,1,* Qian Zhang,2 Wenhui Li,2 Yanchun Ding1 1Department of Cardiology, The Second Hospital of Dalian Medical University, Dalian, Liaoning, People’s Republic of China; 2The Second Hospital of Dalian Medical University, Dal...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Dove Medical Press
2025-07-01
|
| Series: | Journal of Inflammation Research |
| Subjects: | |
| Online Access: | https://www.dovepress.com/identifying-and-diagnosing-lytic-cell-death-genes-in-atherosclerosis-u-peer-reviewed-fulltext-article-JIR |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Guolin Zhang,1,* Ruicong Ma,2,* Hongjin Jin,1,* Qian Zhang,2 Wenhui Li,2 Yanchun Ding1 1Department of Cardiology, The Second Hospital of Dalian Medical University, Dalian, Liaoning, People’s Republic of China; 2The Second Hospital of Dalian Medical University, Dalian, Liaoning, People’s Republic of China*These authors contributed equally to this workCorrespondence: Yanchun Ding, Email yanchunding@dmu.edu.cnBackground: Lytic cell death (LCD) is gaining research attention in chronic inflammatory diseases such as atherosclerosis (AS). Our study investigates the role and mechanism of LCD in AS using machine learning and bioinformatics.Methods: We sourced gene expression data and single-cell sequencing from the GEO database. Differential analysis identified differentially expressed genes (DEGs), which were then intersected with LCD-related genes to determine LCD-associated DEGs (LCDEGs). Machine learning was used to screen characteristic LCDEGs, and an artificial neural network (ANN) model was developed. The diagnostic accuracy of the model was assessed using ROC curves.Results: The results demonstrated that the ANN model possesses a robust diagnostic ability in distinguishing between normal and AS cases, as well as identifying early and advanced stages. Unique AS subtypes were identified using a consensus clustering method. Two subtypes, C1 (non-immune subtype) and C2 (immune subtype), were delineated based on immune landscape analysis and gene set variation analysis functional enrichment. The chi-square test revealed that C1 was linked to early-stage (low-risk) atherosclerotic plaques, whereas C2 was associated with advanced-stage (high-risk) atherosclerotic plaques. At the single-cell level, LCDEG activity was calculated using AUCell and AddModuleScore. LCDEGs exhibited increased activity levels in macrophages within the initially classified cell subtypes. Moreover, they displayed higher activity in the “inflammation” subtype of specific macrophage subtype analysis.Conclusion: This study highlights the clinical potential of LCD in AS and suggests it involves a macrophage-mediated mechanism. We also experimentally identified and validated cytochrome B-245β chain (CYBB) as a potential biomarker for AS. Keywords: atherosclerosis, lytic cell death, lytic cell death-related genes, machine learning, bioinformatics, cytochrome B-245β chain, CYBB |
|---|---|
| ISSN: | 1178-7031 |