Combined analysis of bulk RNA and single-cell RNA sequencing to identify pyroptosis-related markers and the role of dendritic cells in chronic obstructive pulmonary disease

Huiyan Zheng ¹, Guifeng Wang ¹, Yunlai Wang ¹

⁰Department of Health Management Center, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.

Heliyon +

|

March 21, 2024

View abstract on PubMed

Summary

Machine learning identified pyroptosis-related genes for diagnosing chronic obstructive pulmonary disease (COPD). Dendritic cell activity and communication pathways are critical in COPD progression.

Area Of Science

Immunology
Computational Biology
Pulmonology

Background

Chronic obstructive pulmonary disease (COPD) is a progressive lung disease characterized by airflow limitation and dyspnea.
Pyroptosis, a form of programmed cell death, particularly via the NLRP3 inflammasome pathway, is implicated in COPD exacerbations.

Purpose Of The Study

To identify diagnostic pyroptosis-related genes for COPD using machine learning.
To explore the immune landscape and cell-cell communication in COPD.

Main Methods

Machine learning algorithms (deep neural networks, logistic regression) were applied to GEO datasets to screen for pyroptosis-related genes.
External datasets were used for validation.
Single-cell RNA sequencing (scRNA-seq) data analyzed immune cell infiltration and intercellular communication patterns.

Main Results

Deep neural networks and logistic regression achieved high AUC values (0.91 and 0.74) for COPD diagnosis.
Significant differences in dendritic cell (DC) infiltration were observed between COPD patients and controls.
Key pathways mediating communication between DCs and other cells were identified as critical in COPD.

Conclusions

Pyroptosis-related genes identified through machine learning can aid in COPD diagnosis.
Dendritic cell activity and their communication networks play a crucial role in the pathogenesis and progression of COPD.

Keywords:

Biomarkers Chronic obstructive pulmonary disease Dendritic cell Diagnostic model Immune infiltration Machine learning