基于siRNA-mRNA结合热力学特征的高效siRNA筛选  

Selecting highly effective siRNAs by thermodynamics of siRNA-mRNA binding

作　　者：刘元宁[1,2] 徐宝林[1,2] 张浩[1,2] 陈竟博[1,2] 韩烨[1,2] 禹剑龙 

机构地区：[1]吉林大学计算机科学与技术学院,长春130012 [2]吉林大学符号计算与知识工程教育部重点实验室,长春130012

出　　处：《吉林大学学报（工学版）》2014年第1期191-195,共5页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(60971089);吉林大学'大学生创新创业训练计划'创新训练项目(2012B53146)

摘　　要：siRNA与靶标mRNA的相互作用的热力学特征对于其干扰效率至关重要。RNAup是维也纳RNA软件包提供的预测靶标区域的可接触性(accessibility)的软件,将其预测的热力学特征加入到siRNA序列特征里,用支持向量机筛选高效的siRNA,该模型能够很好地预测出高效的siRNA,准确率达到88.12%,敏感度和特异性分别为88.35%和87.94%。The thermodynamic of interactions between siRNAs and target mRNAs is crucial for the efficiency of its interference. RNAup is a software provided by Vienna RNA package, which can predict the accessibility of target region. By adding its predicted thermodynamic characteristics to the SiRNA sequence features and screening the efficient siRNA with support vector machines, the model can well predict efficient siRNA that the accuracy rate is up to 88.12%, the sensitivity and specificity are up to 88.35% and 87.94%, respectively.

关 键 词：计算机应用 热力学 RNAI 支持向量机 干扰效率预测 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]