mRNA-miRNA bipartite networks reconstruction in different tissues of bladder cancer based on gene co-expression network analysis

Abstract

Bladder cancer (BC) is one of the most important cancers worldwide, and if it is diagnosed early, its progression in humans can be prevented and long-term survival will be achieved accordingly. This study aimed to identify novel micro-RNA (miRNA) and gene-based biomarkers for diagnosing BC. The microarray dataset of BC tissues (GSE13507) listed in the GEO database was analyzed for this purpose. The gene expression data from three BC tissues including 165 primary bladder cancer (PBC), 58 normal looking-bladder mucosae surrounding cancer (NBMSC), and 23 recurrent non-muscle invasive tumor tissues (RNIT) were used to reconstruct gene co-expression networks. After preprocessing and normalization, deferentially expressed genes (DEGs) were obtained and used to construct the weighted gene co-expression network (WGCNA). Gene co-expression modules and low-preserved modules were extracted among BC tissues using network clustering. Next, the experimentally validated mRNA-miRNA interaction information were used to reconstruct three mRNA-miRNA bipartite networks. Reactome pathway database and Gene ontology (GO) was subsequently performed for the extracted genes of three bipartite networks and miRNAs, respectively. To further analyze the data, ten hub miRNAs (miRNAs with the highest degree) were selected in each bipartite network to reconstruct three bipartite subnetworks. Finally, the obtained biomarkers were comprehensively investigated and discussed in authentic studies. The obtained results from our study indicated a group of genes including PPARD, CST4, CSNK1E, PTPN14, ETV6, and ADRM1 as well as novel miRNAs (e.g., miR-16-5p, miR-335-5p, miR-124-3p, and let-7b-5p) which might be potentially associated with BC and could be a potential biomarker. Afterward, three drug-gene interaction networks were reconstructed to explore candidate drugs for the treatment of BC. The hub miRNAs in the mRNA-miRNA bipartite network played a fundamental role in BC progression; however, these findings need further investigation.

Figure 1

Bipartite mRNA-miRNA subnetwork for NBMSC-PBC. Cytoscape v.3.8.2 was used to visualize the network.

Figure 2

Bipartite mRNA-miRNA subnetwork for PBC-RNIT. Cytoscape v.3.8.2 was used to visualize the network.

Figure 3

Bipartite mRNA-miRNA subnetwork for NBMSC-RNIT. Cytoscape v.3.8.2 was used to visualize the network.

Figure 4

Venn diagram for the obtained genes and miRNAs. (a) Venn diagram of genes (b) Venn diagram of miRNAs. The webtool (https://bioinformatics.psb.ugent.be/webtools/Venn) was used to construct Venn diagram.

Figure 5

The candidate drug-gene network extracted from DGIdb database. The candidate drugs (lightblue hexagon) identified as regulators of the NBMSC-PBC subnetwork. Among these drugs, some drugs can regulate more than two gene (red hexagon). Cytoscape v.3.8.2 was used to visualize the network.

Figure 6

The candidate drug-gene network extracted from DGIdb database. The candidate drugs (lightblue hexagon) identified as regulators of the NBMSC-RNIT subnetwork. Among these drugs, some drugs can regulate more than two gene (red hexagon). Cytoscape v.3.8.2 was used to visualize the network.

Figure 7

The candidate drug-gene network extracted from DGIdb database. The candidate drugs (lightblue hexagon) identified as regulators of the PBC-RNIT subnetwork. Among these drugs, some drugs can regulate more than two gene (red hexagon). Cytoscape v.3.8.2 was used to visualize the network.