前列腺癌相关lincRNAs功能的生物信息学分析

发布时间：2018-10-23 18:16

【摘要】：长链非编码RNA（long non-coding RNA,lncRNA）是一类长度大于200核苷酸、不具备蛋白编码能力的转录本。目前为止，了解确切功能的lncRNA数量极少。前列腺癌症是发生在男性生殖系统中的一种恶性肿瘤，它也是男性最重要的致死肿瘤之一。随着癌症转录组研究的进展，越来越多的证据显示，lncRNA的异常表达参与了肿瘤的发生。故而lncRNA的研究十分重要。根据在基因组位置的不同，lncRNA可分为不同的亚类，其中基因间的lncRNA则转录自基因间区域，即长链非编码间区RNA（Long intergenic non-coding RNAs, lincRNAs）是本文的探索的对象。我们使用下一代测序（NGS）的数据，共计30个RNA-Seq样本（20个癌症样本，10个癌旁正常组织），使用加权基因共表达网络分析（weighted gene co-expression networkanalysis，WGCNA）的方法对差异表达的蛋白编码基因和lincRNA数据进行共表达分析。我们对这5000个高度差异基因建立共表达scale-free网络并进行聚类分析，获得9个共表达模块，分别命名为M1~M9。我们对表达矩阵进行主成分分析（principalcomponent analysis，PCA），并将第一主成分作为整个模块的eigengene，然后用eigengene的表达值代表整个模块的总体表达。计算了每个基因与eigengene的Pearson’s correlation，作为此基因的的模块关联系数（即module membership kME）。得到前列腺癌共表达模块后，检验所有模块是否与前列腺癌症相关的显著性，得到三个前列腺癌相关的风险模块M1(blackpcor=0.008219622)；M3(cyanpcor=0.027801588)；M5(magenta pcor=0.001489048)。模块M3(“cyan”)和模块M5(“magenta”)的eigengenes的聚类分析结果表明这两个模块接近。我们对这三个前列腺癌风险模块（M1, M3,M5）进行子网络分析，发现在M3，M5模块中lincRNA含量较多。另外在eigengenes的转录因子（transcription factor，TF）富集分析结果表明M3和M5模块共同被重要的癌症调控因子Sp1调控，miRNAs富集分析结果表明模块分别和miR-24和miR-330显著相关(p_value=0.05)，根据相关已报道文献可知miR-24和miR-330是前列腺癌症的的生物标志物。在对M3模块网络的研究中发现，前列腺癌风险模块M3（“cyan”）中有7个lincRNA (TCONS_l2_00008237、 TCONS_l2_00022670、TCONS_l2_00022611、 linc-PXN-1、 TCONS_l2_00011130、 TCONS_l2_00001418、TCONS_l2_00013175)起着重要调控基因共表达的作用，，故而在网络的层次上又验证了lincRNA的生物学功能，我们把这个模块中的eigengene基因在MetaCore GeneGO中进行富集分析，发现M3（”cyan”）模块内的基因主要参与转录调控，细胞内蛋白运输定位，基因表达，衰变等生物过程。所有的结果皆证明了lincRNA在前列腺癌中行使相关调控的功能。本课题通过一系列的生物信息学方法，阐明了lincRNA如何同蛋白编码基因共表达，并行使相关特定的功能；可作为指导性资源，为研究者研究lncRNA提供新的思路和方式。
[Abstract]:Long chain non-coding RNA (long non-coding RNA,lncRNA) is a class of transcripts with a length greater than 200 nucleotides and no protein coding ability. So far, the number of lncRNA that knows the exact function is very small. Prostate cancer is a malignant tumor occurring in the male reproductive system. It is also one of the most important fatal tumors in men. With the development of cancer transcriptome research, there is more and more evidence that abnormal expression of lncRNA is involved in tumorigenesis. Therefore, the study of lncRNA is very important. According to the position of genome, lncRNA can be divided into different subclasses, in which the intergenic lncRNA is transcribed from the intergenic region, that is, the long chain noncoding region (RNA (Long intergenic non-coding RNAs, lincRNAs) is the object of this study. We use the next generation of sequenced (NGS) data, A total of 30 RNA-Seq samples (20 cancer samples and 10 adjacent normal tissues) were used to analyze the differentially expressed protein coding genes and lincRNA data by weighted gene coexpression network analysis (weighted gene co-expression networkanalysis,WGCNA). We established a coexpression scale-free network for the 5 000 highly differentially expressed genes and conducted cluster analysis to obtain 9 coexpression modules named M1 M9 respectively. The expression matrix is analyzed by principal component analysis (principalcomponent analysis,PCA), and the first principal component is taken as the eigengene, of the whole module, and then the expression value of eigengene is used to represent the total expression of the whole module. The correlation coefficient of each gene with the Pearson's correlation, of eigengene as the module of this gene (that is, module membership kME).) is calculated. After the prostate cancer co-expression module was obtained, all the modules were tested for prostate cancer-related significance, and three prostate cancer-related risk modules M1 (blackpcor=0.008219622), M3 (cyanpcor=0.027801588) and M5 (magenta pcor=0.001489048) were obtained. The clustering analysis of M3 ("cyan") and M5 ("magenta") eigengenes shows that the two modules are close to each other. The subnetwork analysis of the three prostate cancer risk modules (M1, M3OM5) showed that the lincRNA content was higher in M3M5 module. In addition, the results of eigengenes transcription factor (transcription factor,TF) enrichment analysis showed that M3 and M5 modules were jointly regulated by the important cancer regulator Sp1. MiRNAs enrichment analysis showed that the modules were significantly correlated with miR-24 and miR-330 (p_value=0.05), respectively, according to the relevant literature. MiR-24 and miR-330 are biomarkers of prostate cancer. In the study of M3 module network, it was found that 7 lincRNA (TCONS_l2_00008237, TCONS_l2_00022670,TCONS_l2_00022611, linc-PXN-1, TCONS_l2_00011130, TCONS_l2_00001418,TCONS_l2_00013175) in M3 ("cyan") play an important role in regulating gene coexpression. Therefore, the biological function of lincRNA was verified at the network level. We enriched the eigengene gene in MetaCore GeneGO and found that the gene in M3 ("cyan") module is mainly involved in transcriptional regulation, intracellular protein transport location, gene expression, decay and other biological processes. All the results suggest that lincRNA plays a role in the regulation of prostate cancer. Through a series of bioinformatics methods, this paper illustrates how lincRNA is co-expressed with protein-encoded genes and performs related specific functions, which can be used as a guiding resource and provide a new way for researchers to study lncRNA.
【学位授予单位】：苏州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R737.25

【共引文献】