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

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.

Figure 1

Comparison between model 1 and model 2. Observed siRNA activities of the Huesken_test are plotted against predicted siRNA activities by model 1 (a) and model 2 (b).

Figure 2

The 57 features selected by the BSFS method.

Figure 3

Boxplots of the top 15 features. For each plot, the left side represents potent siRNAs, and the right side represents nonpotent siRNAs.

Figure 4

ROC curves of the five algorithms.

Figure 5

Comparisons of ten algorithms using the three independent datasets of Vickers, Reynolds, and Harborth.

Algorithm 1

The calculation process of threshold k.