m5CPred-SVM:  A Novel Method for Predicting m5C Sites of RNA

Abstract

Background: As one of the most common post-transcriptional modifications (PTCM) in RNA, 5-cytosine-methylation plays important roles in many biological functions such as RNA metabolism and cell fate decision. Through accurate identification of 5-methylcytosine (m5C) sites on RNA, researchers can better understand the exact role of 5-cytosine-methylation in these biological functions. In recent years, computational methods of predicting m5C sites have attracted lots of interests because of its efficiency and low-cost. However, both the accuracy and efficiency of these methods are not satisfactory yet and need further improvement. Results: In this work, we have developed a new computational method, m5CPred-SVM, to identify m5C sites in three species, H. sapiens, M. musculus and A. thaliana. To build this model, we first collected benchmark datasets following three recently published methods. Then, six types of sequence-based features were generated based on RNA segments and the sequential forward feature selection strategy was used to obtain the optimal feature subset. After that, the performance of models based on different learning algorithms were compared, and the model based on the support vector machine provided the highest prediction accuracy. Finally, our proposed method, m5CPred-SVM was compared with several existing methods, and the result showed that m5CPred-SVM offered substantially higher prediction accuracy than previously published methods. It is expected that our method, m5CPred-SVM, can become a useful tool for accurate identification of m5C sites.Conclusion: In this study, by introducing position-specific propensity related features, we built a new model, m5CPred-SVM, to predict RNA m5C sites of three different species. The result shows that our model outperformed the existing state-of-art models. Our model is available for users through a web server at http://zhulab.ahu.edu.cn/m5CPred-SVM.

M1ARegpred: Epitranscriptome Target Prediction of N1-methyladenosine (m1A) Regulators Based on Sequencing Features and Genomic Features

“…Support vector machine (SVM) is one of the most widely used algorithms [19,51,[54][55][56], which shows stable prediction performance. However, to verify the rationality of choosing SVM as the machine learning algorithm to build our proposed models, the performances of SVM, GBM, random forest, KNN, and bayesian were compared for the prediction of m1A regulator binding sites on the full transcript model and mature mRNA model respectively.…”

Advances in mRNA 5-methylcytosine modifications: Detection, effectors, biological functions, and clinical relevance

“…However, studies also focus on determining optimal composite feature extraction methods and classifiers. 14 Currently, 12 models have been developed to predict mRNA m 5 C modifications: PEA-m 5 C, 117 a hybrid model for quickly and accurately identifying m 5 C sites from non-m 5 C sites in Homo sapiens RNA (Pm 5 CS-Comp-mRMR), 118 m 5 C-PseDNC, 119 iRNA-m 5 C-PseDNC, 120 m 5 C-HPCR, 116 prediction of RNA 5-methylcytosine sites based on three different kinds of nucleotide composition (RNA-m 5 C-Pred), 121 identifying the occurrence sites of different RNA modifications by incorporating collective effects of nucleotides into PseKNC (iRNA-PseColl), 122 RNA-m 5 C-finder, 123 iRNAm 5 C, 119 a novel method for predicting m 5 C sites of RNA (m 5 C-Pred-SVM), 124 a new predictor for multiple types of RNA modification sites using deep learning (DeepMRMP), 125 and a platform for simultaneously identifying multiple kinds of RNA modifications (iMRM) (Table S1). 126 With the exception of PEA-m 5 C, which specifically predicts m 5 C methylation sites in A. thaliana, 104,117 all of these models can predict human m 5 C modifications.…”

“…119 Meanwhile, the iRNA-m 5 C model can detect m 5 C modification sites in four different species (Homo sapiens, Mus musculus, Saccharomyces cerevisiae, and A. thaliana) and has a higher predictive capacity than RNA-m 5 C-finder in terms of precision and accuracy. 119 The m 5 C-Pred-SVM model can detect m 5 C modification sites in H. sapiens, M. musculus, and A. thaliana by introducing position-specific propensity-related features; its performance, including sensitivity, specificity, overall accuracy, and Matthews correlation coefficient, is superior to that of other existing methods 124 . Similarly, DeepMRMP can predict m 1 A, 4, and m 5 C modification sites in H. sapiens, M. musculus, and S. cerevisiae RNA, 125 while integrating multiple modification types in different species, making it both time and cost effective.…”

Analyzing the genome-wide expression of m5C-related regulator genes and prognostic characteristics in pan-cancer

“…Through hypermethylation of the promoter region, DNA m5C-related genes can upregulate gene expression levels and inhibit some tumor suppressor genes. In addition, a m5C-related modi cation of RNA refers to the methylation of the fth C atom of RNA cytosine 4 . An alteration in the m5C-related of mRNA could affect its structure, stability, and translation 5 .…”