Identifying Diagnostic and Prognostic Biomarkers and Candidate Therapeutic Drugs of Gastric Cancer Based on Transcriptomics and Single-Cell Sequencing

Abstract

Background and Objective: Gastric cancer (GC) is an important health burden and the prognosis of GC is poor. We aimed to explore new diagnostic and prognostic indicators as well as potential therapeutic targets for GC in the current study. Methods: We screened the overlapped differentially expressed genes (DEGs) from GSE54129 and TCGA STAD datasets. Protein-protein interaction network analysis recognized the hub genes among the DEGs. The roles of these genes in diagnosis, prognosis, and their relationship with immune infiltrates and drug sensitivity of GC were analyzed using R studio. Finally, the clinically significant hub genes were verified using single-cell RNA sequencing (scRNA-seq) data. Results: A total of 222 overlapping genes were screened, which were enriched in extracellular matrix-related pathways. Further, 17 hub genes were identified, and our findings demonstrated that BGN, COMP, COL5A2, and SPARC might be important diagnostic and prognostic indicators of GC, which were also correlated with immune cell infiltration, tumor mutation burden (TMB), microsatellite instability (MSI), and sensitivity of therapeutic drugs. The scRNA-seq results further confirmed that all four hub genes were highly expressed in GC. Conclusion: Based on transcriptomics and single-cell sequencing, we identified four diagnostic and prognostic biomarkers of GC, including BGN, COMP, COL5A2, and SPARC, which can help predict drug sensitivity for GC as well.

Keywords: bioinformatics; biomarkers; gastric cancer; hub genes; molecular drugs; prognosis.

FIGURE 1

DEGs between GC and normal tissues based on the GEO and TCGA database. Volcano plot of DEGs screened from the GSE54129 microarray dataset (A) and TCGA STAD (B). Red: upregulated; green: downregulated. Overlapping DEGs from the GEO and TCGA database (C); upregulated and downregulated overlapping genes from the GEO and TCGA database (D,E).

FIGURE 2

Scatterplots for GO enrichment and KEGG pathways of DEGs. (A) GO enrichment of DEGs; (B) KEGG pathways of DEGs. The gene ratio is assigned to the x-axis and the description of pathway to the y-axis. The area of the displayed circles is proportional to the number of genes assigned to the term and the color corresponds to the adjusted p-value. (C) GSEA analysis of DEGs. The bar chart displays the normalized enrichment score of pathways that were significantly related to DEGs.

FIGURE 3

The PPI interaction network and co-expression based on the hub genes. (A) The PPI interaction network of the hub genes-coded proteins. Nodes represent proteins and edges represent interactions between two proteins. (B) The heatmap of the co-expression of the hub genes. The darker the color, the stronger the correlation. Asterisks represent levels of significance (*p < 0.05, **p < 0.01).

FIGURE 4

Kaplan-Meier survival curves by the expression level of hub genes. (A), BGN; (B), COL1A2; (C), COL4A1; (D), COL5A1; (E), COL5A2; (F), COL11A1; (G), COMP; (H), SERPINE1; (I), SPARC; and (J), VCAN. The patients were split into high and low expression groups according to the quartile value of the hub gene expression.

FIGURE 5

The Sankey diagram based on four hub genes with clinical and prognostic significance. (A), BGN; (B), COMP; (C), COL5A2; and (D), SPARC. Each column represents a characteristic variable, different colors represent different types, status, or stages, and lines represent the distribution of the same sample in different characteristic variables.

FIGURE 6

Diagnostic value of four hub genes in GC. (A), BGN; (B), COMP; (C), COL5A2; (D), SPARC; (E), multiple-gene comparison analysis using GEPIA. AUC: area under curve; T: tumor; N: normal. The density of color in each block and the number on the right represent the median expression value of a gene in a given tissue, normalized by the maximum median expression value across all blocks.

FIGURE 7

Nomogram based on four hub genes for predicting the probability of 1-, 3-, 5-years OS for GC patients of the TCGA cohort.

FIGURE 8

Correlations between four hub genes’ expression and immune infiltrates in GC. (A), BGN; (B), COMP; (C), COL5A2; and (D), SPARC.

FIGURE 9

Correlations between four hub genes’ expression and TMB/MSI. (A–D), Correlations between four hub genes’ expression and TMB; (E–H), correlations between four hub genes’ expression and MSI. The horizontal axis in the figure represents the expression distribution of the gene, and the ordinate is the expression distribution of the TMB/MSI score.

FIGURE 10

Correlations between four hub genes’ expression and drug sensitivity. The figure shows the top 15 significant drug-gene pairs with significant correlation. X-axis: gene expression; y-axis: drug sensitivity Z scores.

FIGURE 11

Verification results of four hub genes’ expression using scRNA-seq data. (A), BGN; (B), COMP; (C), COL5A2; and (D), SPARC.