Dataset Information

Gastric cancer-associated microRNA expression signatures: integrated bioinformatics analysis, validation, and clinical significance.

ABSTRACT:

Background

Gastric cancer (GC) is one of the common gastrointestinal malignancy worldwide and exhibits a poor prognosis. Increasing studies have indicated that microRNAs play critical roles in the cancer progression and have shown great potential as useful biomarkers. The search for potential diagnostic and prognostic biomarkers of gastric cancer (GC) with integrated bioinformatics analyses has been undertaken in previous studies.

Methods

In this study, the robust rank aggregation (RRA) method was used to perform an integrated analysis of differentially expressed miRNAs (DEMs) from five microarray datasets in the Gene Expression Omnibus (GEO) database to find robust biomarkers for GC. Ultimately, seven miRNAs were filtered from fourteen primary miRNAs using the validation set of The Cancer Genome Atlas (TCGA) database. Based on these results, diagnostic and survival analyses were performed, and logistic regression and Cox regression were used to determine the clinicopathological characteristics of the DEM expression and overall survival.

Results

Nine eligible miRNA datasets related to GC were selected from the GEO database for integrated analysis in this study. Diagnostic analysis implied that these miRNAs could be regarded as promising candidate diagnostic biomarkers in GC tissues, but whether the results of the tissue analysis are consistent with those of peripheral blood analysis requires further validation. The logistic regression indicated that the ectopic expression of these DEMs was relevant to the histological type, anatomical region, and pathological grade of GC. However, the survival and Cox regression analyses suggested that the poor prognosis of GC patients was not strongly dependent on the ectopic expression of the seven miRNAs, but rather, a poor prognosis was associated with age, metastasis, and histological grade.

Conclusions

Based on the results presented in this study it can be concluded that these miRNAs (miR-455-3p, miR-135b-5p, let-7a-3p, miR-195-5p, miR-204-5p, miR-149-5p, and miR-143-3p) might be potential biomarkers for the early diagnosis of GC patients, but this finding should be regarded with caution. A large-scale, prospective, and multicenter cohort study should be performed.

SUBMITTER: Liu X

PROVIDER: S-EPMC8246217 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundClear cell renal cell carcinoma (ccRCC) represents the most invasive and common adult kidney neoplasm. Mounting evidence suggests that microRNAs (miRNAs) are important regulators of gene expression. But their function in tumourigenesis in this tumour type remains elusive. With the development of high throughput technologies such as microarrays and NGS, aberrant miRNA expression has been widely observed in ccRCC. Systematic and integrative analysis of multiple microRNA expression datasets may reveal potential mechanisms by which microRNAs contribute to ccRCC pathogenesis.MethodsWe collected 5 public microRNA expression datasets in ccRCC versus non-matching normal renal tissues from GEO database and published literatures. We analyzed these data sets with an integrated bioinformatics framework to identify expression signatures. The framework incorporates a novel statistic method for abnormal gene expression detection and an in-house developed predictor to assess the regulatory activity of microRNAs. We then mapped target genes of DE-miRNAs to different databases, such as GO, KEGG, GeneGo etc, for functional enrichment analysis.ResultsUsing this framework we identified a consistent panel of eleven deregulated miRNAs shared by five independent datasets that can distinguish normal kidney tissues from ccRCC. After comparison with 3 RNA-seq based microRNA profiling studies, we found that our data correlated well with the results of next generation sequencing. We also discovered 14 novel molecular pathways that are likely to play a role in the tumourigenesis of ccRCC.ConclusionsThe integrative framework described in this paper greatly improves the inter-dataset consistency of microRNA expression signatures. Consensus expression profile should be identified at pathway or network level to address the heterogeneity of cancer. The DE-miRNA signature and novel pathways identified herein could provide potential biomarkers for ccRCC that await further validation.

Project description:Background: The systematic expression characteristics and functions of collagen genes in gastric cancer (GC) have not been reported. Through public data integration, combined with bioinformatics analysis, we identified a panel of collagen genes overexpressed in GC. The functions of these genes were analyzed and validated in a GC-related cohort. microRNAs that may potentially target such genes were investigated in vitro. Methods: Four GC-related datasets retrieved from the Gene Expression Omnibus (GEO) were used to extract differentially expressed genes (DEGs) in GC. Functional annotation was performed to identify the potential roles of the identified DEGs. The association of candidate genes involved in the prognosis of GC patients (n=876) was determined using data provided by the Kaplan-Meier-plotter database, The Cancer Genome Atlas Stomach Adenocarcinoma (TCGA-STAD) repository, and a GC-related dataset (GSE15459). The expression characteristics of candidate genes and their associations with clinical parameters were validated in our in-house cohort (n=58). MicroRNAs able to target the identified candidate genes were predicted and confirmed using qRT-PCR, Western blotting, and dual-luciferase reporter assays in vitro. Results: After the integration of four GEO datasets, 76 DEGs were identified. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analysis indicated that these DEGs were significantly enriched in ECM-related functions and pathways. A group of collagen genes was significantly upregulated in the GC tissues and constituted a protein-protein interaction network as important nodes. Some of these collagen genes were closely associated with poor prognosis in GC patients. Overexpression of COL1A1 and COL4A1 was confirmed in our in-house cohort, and this was related to prognosis and certain clinicopathological parameters. We found that microRNA-29c-3p could directly target COL1A1 and COL4A1 in BGC-823 cells. Conclusions: Collagen genes identified in this study were associated with patient prognosis in GC and may represent diagnostic markers or potential therapeutic targets. Aberrant expression of such candidate genes may be induced by microRNA-29c-3p.