Comparison and Integration of Target Prediction Algorithms for microRNA Studies

Abstract

SummarymicroRNAs are short RNA fragments that have the capacity of regulating hundreds of target gene expression. Currently, due to lack of high-throughput experimental methods for miRNA target identification, a collection of computational target prediction approaches have been developed. However, these approaches deal with different features or factors are weighted differently resulting in diverse range of predictions. The prediction accuracy remains uncertain. In this paper, three commonly used target prediction algorithms are evaluated and further integrated using algorithm combination, ranking aggregation and Bayesian Network classification. Our results revealed that each individual prediction algorithm displays its advantages as was shown on different test data sets. Among different integration strategies, the application of Bayesian Network classifier on the features calculated from multiple prediction methods significantly improved target prediction accuracy.

MiR-365 regulates lung cancer and developmental gene thyroid transcription factor 1

“…S6). A chemical inhibitor of TGFβ receptor I kinase, LY364947, reversed the TGFβ-dependent repression of TTF1 RNA and protein expression ( Although none of the available miRNA/target prediction algorithms is perfect, 42 the high false prediction rate of TargetScan seen in this study was somewhat surprising to us. During the course of our study, we had looked up the miRanda-based prediction for potential miRNA binders of TTF-1 3'UTR.…”

miR-183 potentially inhibits NF-κB1 expression by directly targeting its 3′-untranslated region

“…A collection of miRNA prediction algorithms is currently available that weight different factors, such as seed region pairing, evolutionary conservation, and thermodynamic stability of the mRNA/miRNA duplex, thus generating a diverse range of predictions [16]. While it is hard to determine which algorithm is the most reliable or accurate, both TargetScan and PicTar provide high sensitivity and specificity and have produced experimentally verifiable data sets [17].…”

Endogenous rice ( Oryza Sativa ) miRNAs and their potential targets against Rice Tungro Virus using various string matching algorithms

“…The results indicate that psRNATarget is well capable of performing highthroughput analysis for large-scale datasets. Zhang Y et al [12] mentioned about the performance of prediction accuracy using several target prediction algorithms like miRanda, TargetScan RNAhybrid and of a selection of integration strategies on these algorithms using multiple data sets. miRNA is to measure the performance of miRNA target prediction algorithms using both the true-positive and false-positive rate and a Bayesian Network classifier is used for better target prediction.…”