MatPred: Computational Identification of Mature MicroRNAs within Novel Pre-MicroRNAs

Abstract

Background. MicroRNAs (miRNAs) are short noncoding RNAs integral for regulating gene expression at the posttranscriptional level. However, experimental methods often fall short in finding miRNAs expressed at low levels or in specific tissues. While several computational methods have been developed for predicting the localization of mature miRNAs within the precursor transcript, the prediction accuracy requires significant improvement. Methodology/Principal Findings. Here, we present MatPred, which predicts mature miRNA candidates within novel pre-miRNA transcripts. In addition to the relative locus of the mature miRNA within the pre-miRNA hairpin loop and minimum free energy, we innovatively integrated features that describe the nucleotide-specific RNA secondary structure characteristics. In total, 94 features were extracted from the mature miRNA loci and flanking regions. The model was trained based on a radial basis function kernel/support vector machine (RBF/SVM). Our method can predict precise locations of mature miRNAs, as affirmed by experimentally verified human pre-miRNAs or pre-miRNAs candidates, thus achieving a significant advantage over existing methods. Conclusions. MatPred is a highly effective method for identifying mature miRNAs within novel pre-miRNA transcripts. Our model significantly outperformed three other widely used existing methods. Such processing prediction methods may provide important insight into miRNA biogenesis.

An Approach to Identify Individual Functional Single Nucleotide Polymorphisms and Isoform MicroRNAs

“…Similarly, the training datasets of P3_5 and P3_3 were constructed. Feature set extraction, feature set selection, and classifier training were performed as described for matpred [33]. Specifically, for P5_5, P5_3, P3_5, and P3_3, the position deviation predicted accuracies of the first candidate are 79%, 71%, 66%, and 90%, respectively within 5nt distances.…”

“…In our previous study, we developed matpred [33] to identify mature miRNAs from pre-miRNAs using biological characteristics of the 5′ arm start sites. As is known, isomiRs are always generated from 3′ arm heterogeneity [8, 10].…”

“…In our previous study, we developed software to predict mature miRNAs [32, 33] from a novel pre-miRNA, and the 1000 Genomes Project contains the genomes of 2504 individuals from 26 populations. It includes more than 88 million variants, 3.6 million short insertions/deletions, and 60,000 structural variants, and the distribution of genetic variations across the global sample plays important roles in disease [34].…”

Adaboost-SVM-based probability algorithm for the prediction of all mature miRNA sites based on structured-sequence features

“…As a result, the method finds that 7, 8 and 9 nt from the starting position have the typical biological features to distinguish mature miRNAs. Microprocessor SVM 12 , MiRpara 13 , MaturePred 14 and matPred 15 are developed using the SVM algorithm. Microprocessor SVM is proposed based on 686 features that are associated with sequence and structure, the accuracy of this method is 50%, and 90% of its predictions are within a 2 nt deviation.…”

Improving classification of mature microRNA by solving class imbalance problem

“…Therefore, mature miRNA identification problem should be also considered as imbalance class distribution. Several computational methods which have been developed for mature miRNAs classification include MatureByes 21 , MiRMat 22 , MiRRim2 23 , MiRdup 24 , MaturePred 25 , MiRPara 26 , mirExplorer 27 , Matpred 28 and MiRduplexSVM 29 . These methods mainly focus on the use of machine learning techniques, and their lower predictive performance suffers from the class-imbalance problem.…”