Table of Contents Author Guidelines Submit a Manuscript
Computational and Mathematical Methods in Medicine
Volume 2017, Article ID 5043984, 12 pages
https://doi.org/10.1155/2017/5043984
Research Article

Utilizing Selected Di- and Trinucleotides of siRNA to Predict RNAi Activity

Ye Han,1,2 Yuanning Liu,1,2 Hao Zhang,1,2 Fei He,3,4,5 Chonghe Shu,1,2 and Liyan Dong1,2

1Department of Computer Science and Technology, Jilin University, Changchun, Jilin, China
2Key Laboratory of Symbolic Computation and Knowledge Engineering, Ministry of Education, Jilin University, Changchun, China
3Department of Computer Science and Information Technology, Northeast Normal University, Changchun, Jilin, China
4Department of Environment, Northeast Normal University, Changchun, Jilin, China
5Institute of Computational Biology, Northeast Normal University, Changchun, China

Correspondence should be addressed to Liyan Dong; nc.ude.ulj@ylgnod

Received 25 October 2016; Accepted 15 December 2016; Published 24 January 2017

Academic Editor: Yu Xue

Copyright © 2017 Ye Han et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Small interfering RNAs (siRNAs) induce posttranscriptional gene silencing in various organisms. siRNAs targeted to different positions of the same gene show different effectiveness; hence, predicting siRNA activity is a crucial step. In this paper, we developed and evaluated a powerful tool named “siRNApred” with a new mixed feature set to predict siRNA activity. To improve the prediction accuracy, we proposed 2-3NTs as our new features. A Random Forest siRNA activity prediction model was constructed using the feature set selected by our proposed Binary Search Feature Selection (BSFS) algorithm. Experimental data demonstrated that the binding site of the Argonaute protein correlates with siRNA activity. “siRNApred” is effective for selecting active siRNAs, and the prediction results demonstrate that our method can outperform other current siRNA activity prediction methods in terms of prediction accuracy.