Identification of a 5‑microrna Signature and Hub Mirna‑mrna Interactions Associated with Pancreatic Cancer.

Xianxiong Ma,Ruikang Tao,Lei Li,Hengyu Chen,Zongtao Liu,Jie Bai,Xiaoming Shuai,Chuanqing Wu,Kaixiong Tao
DOI: https://doi.org/10.3892/or.2018.6820
2019-01-01
Oncology Reports
Abstract:miRNA‑gene axes have been reported to serve an important role in the carcinogenesis of pancreatic cancer (PC). The aim of the present study was to systematically identity the microRNA signature and hub molecules, as well as hub miRNA‑gene axes, and to explore the potential biomarkers and mechanisms associated with the carcinogenesis of PC. Eleven microRNA profile datasets were obtained from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) and ArrayExpress databases, and a meta‑analysis was performed to identify the differentially expressed miRNAs (DEMs) between tumor tissue and normal tissue. Subsequently, a diagnostic regression model was constructed to identify PC based on The Cancer Genome Atlas (TCGA) miRNA sequence data by using the least absolute shrinkage and selection operator (LASSO) method. In addition, GSE41368 was downloaded, and a weighted gene co‑expression network analysis (WGCNA) was performed to obtain the gene module associated with carcinogenesis by using the TCGAbiolinks and WGCNA packages, respectively. Finally, miRNA‑gene networks were constructed and visualized using Cytoscape software, followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses based on the Database for Annotation, Visualization, and Integrated Discovery (DAVID). A total of 14 DEMs were identified, and a 5‑microRNA‑based score generated by the LASSO regression model provided a high accuracy for identifying PC [area under the curve (AUC)=0.918]. In addition, 44 miRNA‑mRNA interactions were constructed, and 4 hub genes were screened on the basis of the above bioinformatic tools and databases. Furthermore, 14 biological process (BP) functions and 6 KEGG pathways were identified according to gene set enrichment analysis (GSEA). In summary, the present study applied integrated bioinformatics approaches to generate a holistic view of PC, thereby providing a basis for further clinical application of the 5‑miRNA signature and the identified hub molecules, as well as the miRNA‑gene axes, which could serve as diagnostic markers and potential treatment targets.
What problem does this paper attempt to address?