# Based on single-cell and transcriptome analysis of inflammatory pathway biomarkers and their molecular mechanisms in chronic obstructive pulmonary disease

## TL;DR

This study identifies key genes and cell types involved in COPD inflammation using single-cell and transcriptome data, offering new insights for precision therapies.

## Contribution

The novel integration of transcriptomic and single-cell analyses reveals key genes and macrophage dynamics in COPD inflammation.

## Key findings

- CXCL12, CXCR4, GGT1, and VWF are key genes in COPD pathogenesis.
- Macrophages are the only cell type with significant abundance differences in COPD.
- CXCR4 shows persistent expression throughout macrophage differentiation.

## Abstract

Systemic inflammation in chronic obstructive pulmonary disease (COPD) presents significant therapeutic challenges. Our study employs integrated transcriptomic and single-cell analyses to identify inflammation-related biomarkers and elucidate their pathogenic mechanisms in COPD.

Training dataset GSE37768, validation dataset GSE239897, and single-cell dataset GSE249584 were retrieved from the GEO database. Inflammation-associated genes were screened from the GeneCards database. Differential expression analysis was employed to identify candidate genes, followed by machine learning approaches and expression validation to pinpoint key genes. Functional characterization of these key genes was conducted through Gene Set Enrichment Analysis (GSEA), immune infiltration profiling, molecular regulatory network construction, drug prediction, and GeneMANIA interaction analysis. Single-cell data analysis elucidated cellular heterogeneity and identified critical cell types. Pseudotime analysis was subsequently performed to investigate the roles of key genes throughout developmental trajectories within these critical cell types.

Twelve candidate genes associated with COPD and inflammation were screened, followed by GO and KEGG enrichment analyses. Subsequently, Least Absolute Shrinkage and Selection Operator (LASSO) regression and Support Vector Machine-Recursive Feature Elimination (SVM-RFE) modeling identified six candidate key genes. Among these, only CXCL12, CXCR4, GGT1, and VWF exhibited consistent expression patterns across both training and validation datasets, establishing them as key genes. Their diagnostic value was further validated by constructing an artificial neural network model. Immune infiltration analysis revealed aberrant basophil abundance in COPD. Single-cell analysis annotated 11 distinct cell types, with macrophages representing the sole cell type demonstrating significant abundance differences between COPD and control groups. Pseudotime trajectory analysis delineated nine differentiation states, wherein CXCR4 expression persisted throughout the cellular differentiation trajectory.

This study identified CXCL12, CXCR4, GGT1, and VWF as key genes in COPD pathogenesis. Macrophages constituted the only cell type exhibiting significant abundance alterations, with CXCR4 demonstrating persistent expression throughout macrophage differentiation trajectories. These findings provide valuable insights and suggest potential directions for developing precision therapeutic strategies for COPD.