Exploring the Molecular Mechanisms of Asthma Across Multiple Datasets

Abstract

Objective: The aim of this research was to utilize bioinformatics techniques to explore the molecular mechanisms at the gene level that contribute to asthma, with the objective of discovering new treatment strategies and potential targets for addressing the condition. Methods: The Series Matrix File data files of GSE43696 and GSE67940 were downloaded from the NCBI GEO public database, including expression profile data of 212 patients. Differential gene expression was functionally annotated using clusterProfiler to evaluate relevant functional categories with GO and KEGG. A gene co-expression network was constructed using MEGENA, and feature importance was evaluated by random forest algorithm. Fluorescent quantitative PCR was employed to validate the expression of essential genes, and the variations in KEGG signaling pathways among the groups with high and low expression were examined through GSEA. Asthma targeted therapeutic drugs were predicted using The Connectivity Map. Finally, single-cell sequencing data were annotated and analyzed using the Seurat and celldex packages. Results: This study screened 267 differentially expressed genes between asthma patients and healthy controls from the GSE43696 dataset and further analyzed them using pathway analysis and multi-scale embedded gene co-expression network analysis, ultimately selecting 12 genes as the candidate gene set for random forest analysis. Based on this, five key genes were selected using random forest algorithm, and their expression was validated in the external dataset GSE67940. The expression of C1orf64 and C7orf26 genes was found to be different between the two groups of patients, and these two genes were found to be associated with immune regulatory factors, chemokines, and cell receptors. The mRNA expression levels of C1orf64 and C7orf26 were consistent with the results of the screening by PCR. Further analysis showed that C1orf64 and C7orf26 were enriched in ABC transporters, cell cycle, cell adhesion molecules, and Notch signaling pathways, and were related to other genes related to asthma. Finally, by classifying the differentially expressed genes using the Connectivity Map, potential clues were provided for finding candidate drugs for asthma treatment. Conclusion: This study combined bioinformatics methods to identify key genes and pathways for asthma. C1orf64 and C7orf26 genes may be the core genes in the pathogenesis of asthma in asthma patients compared to healthy controls, providing potential targets for asthma treatment. These results also suggest the potential application of drug prediction analysis using CMap and single-cell sequencing analysis in understanding the molecular mechanisms of asthma.