Single-cell RNA-seq analysis to identify potential biomarkers for diagnosis, and prognosis of non-small cell lung cancer by using comprehensive bioinformatics approaches

Abstract

Non-small cell lung cancer (NSCLC) is the most common type of lung cancer and the leading cause of cancer-related deaths worldwide. Identification of gene biomarkers and their regulatory factors and signaling pathways is very essential to reveal the molecular mechanisms of NSCLC initiation and progression. Thus, the goal of this study is to identify gene biomarkers for NSCLC diagnosis and prognosis by using scRNA-seq data through bioinformatics techniques. scRNA-seq data were obtained from the GEO database to identify DEGs. A total of 158 DEGs (including 48 upregulated and 110 downregulated) were detected after gene integration. Gene Ontology enrichment and KEGG pathway analysis of DEGs were performed by FunRich software. A PPI network of DEGs was then constructed using the STRING database and visualized by Cytoscape software. We identified 12 key genes (KGs) including MS4A1, CCL5, and GZMB, by using two topological methods based on the PPI networking results. The diagnostic, expression, and prognostic potentials of the identified 12 key genes were assessed using the receiver operating characteristics (ROC) curve and a web-based tool, SurvExpress. From the regulatory network analysis, we extracted the 7 key transcription factors (TFs) (FOXC1, YY1, CEBPB, TFAP2A, SREBF2, RELA, and GATA2), and 8 key miRNAs (hsa-miR-124-3p, hsa-miR-34a-5p, hsa-miR-21-5p, hsa-miR-155-5p, hsa-miR-449a, hsa-miR-24-3p, hsa-let-7b-5p, and hsa-miR-7-5p) associated with the KGs were evaluated. Functional enrichment and pathway analysis, survival analysis, ROC analysis, and regulatory network analysis highlighted crucial roles of the key genes. Our findings might play a significant role as candidate biomarkers in NSCLC diagnosis and prognosis.

Keywords: Diagnosis; Gene biomarkers; Networking analysis; Non-small cell lung cancer; Single-cell RNA-sequencing.

Graphical abstract

Fig. 1

(A) The heatmap of the top 10 marker genes of each cluster where each row represents genes and column represents clusters. (B) Trajectory plot differentiated by eight clusters.

Fig. 2

Top GO and KEGG terms enriched by DEGs ((A) Biological processes, (B) Cellular components, (C) Molecular functions, and (D) Kyoto encyclopedia of genes and genome (KEGG)).

Fig. 3

Protein-protein interaction (PPI) network of DEGs. The green and yellow colors indicate up- and down-regulated genes respectively. The key genes (KGs) are highlighted in diamond shape.

Fig. 4

(A) Kaplan-Meier plot displaying the prognostic effect of the KGs on NSCLC. (B) Boxplot displaying the expression pattern of KGs between risk groups. Red indicates high-risk group and green indicates low-risk group.

Fig. 5

ROC curve evaluating the diagnostic performance of the KGs in NSCLC. Red color indicates GSE75037 dataset and green color indicates GSE19188.

Fig. 6

Gene-miRNAs and Gene-TFs interaction network. (A) In the Gene-miRNAs interaction network, the red bigger circles indicate genes and the blue squares indicate miRNAs. (B) In Gene-TFs interaction network, the red circles indicate genes and the blue diamond shape indicates TFs.