Identification of hub genes between moderate to severe asthma and early lung adenocarcinoma through bioinformatics analysis

Abstract

The objective of this study was to explore the genetic link between moderate to severe asthma and early-stage lung adenocarcinoma (LUAD) using bioinformatic methods. The Cancer Genome Atlas gene-expression profiles for early-stage LUAD and GSE76225 data set for moderate to severe asthma were selected for weighted gene co-expression network analysis, and intersected with the relevant module genes and selected hub genes; the relevant network of hub genes was then determined through a protein-protein interaction network. In addition, gene-set enrichment analysis and gene-set variation analysis (GSVA) were conducted on differentially expressed genes between normal and tumor groups. Gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway-enrichment analyses were applied to detect hub gene-related biological functions. Receiver operating characteristic (ROC) curves were employed to confirm the diagnostic value of hub genes. We identified four key genes, of which SFTPC exhibited relatively high value for areas under the ROC curves, indicating high diagnostic value for moderate to severe asthma. The clinical efficacy of SFTPC was thus consistent with GSVA results, indicating that moderate to severe asthma can inhibit the occurrence of early LUAD.

Keywords: Diagnostic value; Early-stage lung adenocarcinoma; Enrichment analysis; Moderate to severe asthma; Protein–protein interaction network; Weighted gene co-expression network analysis.

Fig. 1

Sample dendrogram and clinical trait heatmap. (A) The sample clustering and clinical traits of early-stage LUAD. (B) The sample clustering and clinical traits of moderate to severe asthma. Clinical characteristic values were converted into continuous colors, with white indicating low and red indicating high.

Fig. 2

Soft-thresholding power and mean connectivity. (A, B) The soft-thresholding power of early-stage LUAD was 9. (C, D) The soft-thresholding power of moderate to severe asthma was 2.

Fig. 3

Gene cluster tree and module diagram. (A) Module cluster dendrogram of early-stage LUAD. (B) Gene cluster tree and its corresponding modules of early-stage LUAD. (C) Module cluster dendrogram of moderate to severe asthma. (D) Gene cluster tree and its corresponding modules of moderate to severe asthma.

Fig. 4

The relationship between modules and clinical traits. (A) In the brown module, the correlation coefficient was − 0.69, and the p-value was 1e−41. (B) In the black module, the correlation coefficient was 0.98, and the p-value was 4e−61. (C) The module membership and gene significance of the brown module. (D) The module membership and gene significance of the black module.

Fig. 5

Volcanic map of DEGs distribution. Red dot: upregulated expression; Blue dots: downregulated expression; Gray dots: meaningless expressions.

Fig. 6

GSEA and GSVA results. (A) CXCR4-Gnαq-PLCβ. (B) PRNP-PI3K-NOX2. (C) RTK-RAS-ERK. (D) TLR2/4-MAPK. (E) SOD1. (F) Heat map of GSVA analysis results.

Fig. 7

Venn diagram and PPI network. (A) Venn diagram of the brown and black module genes. (B) Reverse cumulative distribution curve of residues. (C) ROC curve showing the diagnostic capability of eight machine learning model. (D) Gene Venn diagram of machine learning models. (E) Overlapping gene PPI network.

Fig. 8

GO and KEGG enrichment analysis. (A) Biological process. (B) Cellular components. (C) Molecular function. (D) KEGG.

Fig. 9

ROC of hub genes. (A–D) ROC curves fo SFTC, SMAD6, FEZ1 and GPRC5A. (E–H) AUC plots of fivefold cross-validation.