Characterization and Stratification of LncRNA Modulation Networks

• Main Authors: Man-Ju Hsu, 許嫚茹
• Other Authors: Hsuan-Cheng Huang
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/56044323219063377121

碩士 === 國立陽明大學 === 生物醫學資訊研究所 === 102 === Recent studies found that non-coding RNAs represent most of the human transcriptome and were often dysregulated in cancer. A major family among them is the long non-coding RNA (lncRNA), defined as the non-protein coding transcripts longer than 200 nucleotides to distinguish from small regulatory RNAs such as microRNAs (miRNAs). Like mRNAs, lncRNAs contain miRNA-binding sites and can communicate with and regulate each other by competing specifically for shared miRNAs, thus acting as competing endogenous RNAs (ceRNAs). Understanding the crosstalk among these RNA transcripts will lead to significant insight into gene regulatory networks and have implications in human disease. In this study, we collected the lncRNA, mRNA, and miRNA expression profiles of glioblastoma multiforme from The Cancer Genome Atlas, and developed a novel integrative network analysis approach to construct the modulation networks in diverse cancer clusters and elucidate their regulatory complexity. We first applied a similarity network fusion method to aggregate the three types of RNA data and stratify the cancer clusters. Next, we evaluated several association index methods to quantify the correlation among lncRNAs, mRNAs, and miRNAs. Finally, we applied a modulatory analysis to infer the ceRNA effects among mRNAs and lncRNAs. The results revealed the network characteristics and stratification of lncRNA and ceRNA modulation in cancer.