Functional Exploration Of Non-coding RNA Genes In Human Malignant Tumors Based On Bioinformatics Analysis

BackgroundEstrogen receptor signalling plays important regulatory roles in multiple mammalian physiological processes.Dysregulation of estrogen receptor(ER)expression and/or its associated signalling pathway is strongly associated with the development,progression,transition,and endocrine-resistance of breast cancer.Non-coding transcripts are essential regulators of almost every level of gene regulation.However,few long non-coding transcripts(lncRNAs)associated with the estrogen receptor signalling pathway have been well-described.Material and MethodAll the gene expression microarrys are Affymatrix Hgu133 Plus 2 platform and download from Gene Expression Omnibus database.Microarry probes were annotated with chip annotation files provided by ncFANs.we identified estrogen receptor agitation-related ncRNAs through calculate the differential expressed genes between the test and control groups with Limma R package.Then,we construct coding-noncodig co-expression network to prediect the potential function of ERAR lncRNA.Further,we validate the clinical significance of ERAR lncRNAs in two separate ER positive brest cancer patients trials.ResultWe used array-based methods to identify 33 estrogen receptor agitation-related(ERAR)lncRNAs.A coding-non-coding gene co-expression network analysis suggested that 15 ERAR lncRNAs were associated with mitosis,DNA damage,and DNA repair.Kaplan-Meier analysis indicated that five ERAR lncRNAs selected using the Random Forest-Recursive Feature Elimination algorithm were significantly correlated with endocrine resistance-free survival and distant metastasis-free survival as well as disease free survival.ConclusionOur results suggest that ERAR lncRNAs may serve as novel biomarkers for guiding breast cancer treatment and prognosis.Furthermore,our findings reveal a new avenue by which estrogen receptor signalling can be further explored.Background Both morbidity and mortality of HCC keep ranking the top 3 among tumors in the past 2 decades.Nearly 85 percent of HCC was induced by chonic HBV infection in China.During chronic infection.During the HBV infection,HBV DNA can integrate into the human genome and this has been postulated as a possible mechanism of HBV-induced HCC.Material and Method In this study we used 2199 HBV integration sites from Dr.VIS v2.0 and mapped them to the human genome(hgl9)to obtain viral integration sites(VIS)related to protein-coding and non-protein-coding genes.Then we depict the distribution of the HBV integration on each chromsome in both tumor and non-tumor tissue.Result In total,we found 1377 and 767 VIS within close proximity to protein coding genes and noncoding genes,respectively.Genes affected more than two times included 23.1% of protein-coding genes and 24.7% of long noncoding RNAs(IncRNA).Only 4.8% of VIS were shared between HCC and non-tumor tissues.HBV integrations were more common in chromosomes 5,8,10 and 19 in HCC tissue and chromosomes 1 and 2 in non-tumorous tissue.The number of integration sites on each chromosome correlated with the number of fragile sites in non-tumorous tissue but not in HCC tissue.Functional enrichment analysis of the protein-coding genes containing or in close proximity to HBV integration sites in HCC tissue showed an enrichment of cancer related gene ontology tenns.Additionally,the most frequently associated IncRNA genes were related to telomere maintenance,protein modification processes,and chromosome localization.Conclusion These results shown that HBV integration may have some preference in human chromsome.The distribution of HBV integration in HCC tissue is totally different with non-tumor tissue and some specific HBV integration may contribut to the HCC tumor genesis.