Construction and Analysis of a ceRNA Network Reveals Potential Prognostic Markers in Colorectal Cancer

Abstract

Colorectal cancer (CRC) is one of the leading causes of cancer-related deaths worldwide and is derived from an accumulation of genetic and epigenetic changes. This study explored potential prognostic markers in CRC via the construction and in-depth analysis of a competing endogenous RNA (ceRNA) network, which was generated through a three-step process. First, we screened candidate hub genes in CRC as the primary gene markers to survey their related regulatory non-coding RNAs, miRNAs. Second, the interacting miRNAs were used to search for associated lncRNAs. Thus, candidate RNAs were first constructed into ceRNA networks based on close associations with miRNAs. Further analysis at the isomiR level was also performed for each miRNA locus to understand the detailed expression patterns of the multiple variants. Finally, RNAs were performed an in-depth analysis of expression correlations, which contributed to further screening and validation of potential RNAs with close correlations to each other. Using this approach, nine hub genes, 13 related miRNAs, and 29 candidate lncRNAs were collected and used to construct the ceRNA network. Further in-depth analysis identified the MFAP5-miR-200b-3p-AC005154.6 axis as a potential prognostic marker in CRC. MFAP5 and miR-200b-3p have previously been reported to play important roles in tumorigenesis. These RNAs showed potential prognostic values, and the combination of them may have more sensitivity than using them alone. In conclusion, MFAP5, miR-200b-3p, and AC005154.6 may have potential prognostic value in CRC and may provide a prognostic reference for this patient population.

Identifying Complex lncRNA/Pseudogene-miRNA-mRNA Crosstalk in Hormone-Dependent Cancers

“…Here we investigated the availability of a shared ceRNA network of four common HD cancers. Previous HD cancerrelated ceRNA analyses have focused on lncRNA-associated networks [6][7][8][9][10]. We extended the scope of ceRNA research to include pseudogene-associated cross-HD cancer ceRNA networks.…”

“…Previous bioinformatics studies have identified ceRNA candidates as prognostic or predictive biomarkers for common cancer types such as colorectal, endometrial, prostate, and breast cancers [6][7][8][9][10]. Several web-based tools such as miRTissuece, LncACTdb 2.0, and lnCeDB have been developed, supporting the search for ceRNA interactions networks in multiple tissues [11][12][13].…”

“…Several web-based tools such as miRTissuece, LncACTdb 2.0, and lnCeDB have been developed, supporting the search for ceRNA interactions networks in multiple tissues [11][12][13]. A recent colorectal cancer ceRNA study identified a network of nine hub genes, thirteen lncRNAs, and twenty-nine candidate miRNAs integrating multiple genomic datasets [6]. The authors further revealed the MFAP5-miR-200b-3p-AC005154.6 axis as a potential biomarker of colorectal cancer.…”

Identifying Complex lncRNA/Pseudogene–miRNA–mRNA Crosstalk in Hormone-Dependent Cancers

“…Previous bioinformatics studies have identified ceRNA candidates as prognostic or predictive biomarkers for common cancer types such as colorectal, endometrial, prostate, and breast cancers [6][7][8][9][10]. Several web-based tools such as miRTissue ce , LncACTdb 2.0, and lnCeDB have been developed, supporting the search for ceRNA interaction networks in multiple tissues [11][12][13].…”

“…Several web-based tools such as miRTissue ce , LncACTdb 2.0, and lnCeDB have been developed, supporting the search for ceRNA interaction networks in multiple tissues [11][12][13]. A recent colorectal cancer ceRNA study identified a network of nine hub genes, thirteen lncRNAs, and twenty-nine candidate miRNAs, integrating multiple genomic datasets [6]. The authors further revealed the MFAP5-miR-200b-3p-AC005154.6 axis as a potential biomarker of colorectal cancer.…”