利用miRNA校正的mRNA数据揭示上游基因表达调控因子

Unmasking Upstream Gene Expression Regulators with miRNA-corrected mRNA Data.

作者信息

Bollmann Stephanie, Bu Dengpan, Wang Jiaqi, Bionaz Massimo

机构信息

Department of Integrative Biology, Oregon State University, Corvallis, OR, USA.

State Key Laboratory of Animal Nutrition, Institute of Animal Science, Chinese Academy of Agricultural Sciences, Beijing, China.; CAAS-ICRAF Joint Laboratory on Agroforestry and Sustainable Animal Husbandry, East and Central Asia, World Agroforestry Centre, Beijing, China.

出版信息

Bioinform Biol Insights. 2016 May 29;9(Suppl 4):33-48. doi: 10.4137/BBI.S29332. eCollection 2015.

DOI:10.4137/BBI.S29332

PMID:27279737

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC4886696/

Abstract

Expressed micro-RNA (miRNA) affects messenger RNA (mRNA) abundance, hindering the accuracy of upstream regulator analysis. Our objective was to provide an algorithm to correct such bias. Large mRNA and miRNA analyses were performed on RNA extracted from bovine liver and mammary tissue. Using four levels of target scores from TargetScan (all miRNA:mRNA target gene pairs or only the top 25%, 50%, or 75%). Using four levels of target scores from TargetScan (all miRNA:mRNA target gene pairs or only the top 25%, 50%, or 75%) and four levels of the magnitude of miRNA effect (ME) on mRNA expression (30%, 50%, 75%, and 83% mRNA reduction), we generated 17 different datasets (including the original dataset). For each dataset, we performed upstream regulator analysis using two bioinformatics tools. We detected an increased effect on the upstream regulator analysis with larger miRNA:mRNA pair bins and higher ME. The miRNA correction allowed identification of several upstream regulators not present in the analysis of the original dataset. Thus, the proposed algorithm improved the prediction of upstream regulators.

摘要

表达的微小RNA（miRNA）会影响信使RNA（mRNA）丰度，从而妨碍上游调节因子分析的准确性。我们的目标是提供一种算法来纠正这种偏差。对从牛肝脏和乳腺组织中提取的RNA进行了大规模的mRNA和miRNA分析。使用TargetScan的四个水平的靶标分数（所有miRNA:mRNA靶基因对或仅前25%、50%或75%）。使用TargetScan的四个水平的靶标分数（所有miRNA:mRNA靶基因对或仅前25%、50%或75%）以及miRNA对mRNA表达影响程度（ME）的四个水平（mRNA降低30%、50%、75%和83%），我们生成了17个不同的数据集（包括原始数据集）。对于每个数据集，我们使用两种生物信息学工具进行上游调节因子分析。我们检测到，更大的miRNA:mRNA对分组和更高的ME对上游调节因子分析有增强作用。miRNA校正使得能够识别原始数据集分析中不存在的几个上游调节因子。因此，所提出的算法改进了上游调节因子的预测。