Zhao Jian, Song Xiaofeng, Xu Tianyi, Yang Qichang, Liu Jingjing, Jiang Bin, Wu Jing
Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China.
College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China.
Front Genet. 2021 Jan 13;11:607722. doi: 10.3389/fgene.2020.607722. eCollection 2020.
Increasing lncRNA-associated competing triplets were found to play important roles in cancers. With the accumulation of high-throughput sequencing data in public databases, the size of available tumor samples is becoming larger and larger, which introduces new challenges to identify competing triplets. Here, we developed a novel method, called LncMiM, to detect the lncRNA-miRNA-mRNA competing triplets in ovarian cancer with tumor samples from the TCGA database. In LncMiM, non-linear correlation analysis is used to cover the problem of weak correlations between miRNA-target pairs, which is mainly due to the difference in the magnitude of the expression level. In addition, besides the miRNA, the impact of lncRNA and mRNA on the interactions in triplets is also considered to improve the identification sensitivity of LncMiM without reducing its accuracy. By using LncMiM, a total of 847 lncRNA-associated competing triplets were found. All the competing triplets form a miRNA-lncRNA pair centered regulatory network, in which ZFAS1, SNHG29, GAS5, AC112491.1, and AC099850.4 are the top five lncRNAs with most connections. The results of biological process and KEGG pathway enrichment analysis indicates that the competing triplets are mainly associated with cell division, cell proliferation, cell cycle, oocyte meiosis, oxidative phosphorylation, ribosome, and p53 signaling pathway. Through survival analysis, 107 potential prognostic biomarkers are found in the competing triplets, including FGD5-AS1, HCP5, HMGN4, TACC3, and so on. LncMiM is available at https://github.com/xiaofengsong/LncMiM.
越来越多的长链非编码RNA相关竞争性三联体被发现参与癌症的发生发展过程。随着公共数据库中高通量测序数据的不断积累，可用于研究的肿瘤样本规模越来越大，这为竞争性三联体的识别带来了新的挑战。在此，我们开发了一种名为LncMiM的新方法，用于利用TCGA数据库中的肿瘤样本检测卵巢癌中的长链非编码RNA-微小RNA-信使核糖核酸竞争性三联体。在LncMiM中，非线性相关分析用于解决微小RNA-靶标对之间相关性较弱的问题，这主要是由于表达水平大小的差异所致。此外，除了微小RNA，长链非编码RNA和信使核糖核酸对三联体中相互作用的影响也被考虑在内，以提高LncMiM的识别灵敏度，同时不降低其准确性。通过使用LncMiM，共发现了847个长链非编码RNA相关的竞争性三联体。所有竞争性三联体构成了一个以微小RNA-长链非编码RNA对为中心的调控网络，其中ZFAS1、SNHG29、GAS5、AC112491.1和AC099850.4是连接数最多的前五个长链非编码RNA。生物学过程和KEGG通路富集分析结果表明，竞争性三联体主要与细胞分裂、细胞增殖、细胞周期、卵母细胞减数分裂、氧化磷酸化、核糖体以及p53信号通路相关。通过生存分析，在竞争性三联体中发现了107个潜在的预后生物标志物，包括FGD5-AS1、HCP5、HMGN4、TACC3等。LncMiM可在https://github.com/xiaofengsong/LncMiM获取。
