您的位置:首页  > 论文页面

基于机器学习的miRNA靶标预测方法

发表时间:2008-05-31  浏览量:2100  下载量:818
全部作者: 刘辉,张林,徐钊
作者单位: 中国矿业大学信息与电气工程学院
摘 要: microRNA(miRNA)是一类长约20个核苷酸(nt)的非编码调控小分子RNA.它能够切割靶mRNA或抑制靶mRNA合成蛋白质的过程,因而对后转录调控有非常重大的作用。正确地预测miRNA的靶基因已成为业界研究的热点之一。本文提出一种新的基于统计理念的学习方法进行miRNA靶标判别特征的提取,再使用SVM分类器进行miRNA靶标的预测。仿真结果表明,该方法比目前已有的基于机器学习的miTarget算法的性能更佳。
关 键 词: 生物信息学;微小核糖核酸;支持向量机;靶标预测;分类器
Title: A machine learning approach for miRNA target prediction
Author: LIU Hui, ZHANG Lin, XU Zhao
Organization: School of Information and Electrical Engineering, China University of Mining and Technology
Abstract: MicroRNAs(miRNAs) are a kind of noncoding regulatary RNAs around 20 nucleotides. They are known to possess important post-transcriptional regulatory functions such as slicing mRNAs or repress the sythesization of protein. It is a reserch hot point to predict miRNA targeting correctly. In this paper, a novel machine learning method based on statistic was approched for the discriminative feature extraction for miRNA targeting. Then SVM classifier is adopted for the prediction of miRNA targeting. Finally through simulation, perfomance is shown in comparison with another method of miTarget. And the proposed algorithm can get better prediction results of miRNA targeting.
Key words: bioinformatics; microRNA; support vector machine(SVM); targeting prediction; classifier
发表期数: 2008年9月第10期
引用格式: 刘辉,张林,徐钊. 基于机器学习的miRNA靶标预测方法[J]. 中国科技论文在线精品论文,2008,1(10):1136-1140.
 
0 评论数 0
暂无评论
友情链接