Computational Prediction of Protein Arginine Methylation Based on Composition–Transition–Distribution Features

Abstract

Arginine methylation is one of the most essential protein post-translational modifications. Identifying the site of arginine methylation is a critical problem in biology research. Unfortunately, biological experiments such as mass spectrometry are expensive and time-consuming. Hence, predicting arginine methylation by machine learning is an alternative fast and efficient way. In this paper, we focus on the systematic characterization of arginine methylation with composition–transition–distribution (CTD) features. The presented framework consists of three stages. In the first stage, we extract CTD features from 1750 samples and exploit decision tree to generate accurate prediction. The accuracy of prediction can reach 96%. In the second stage, the support vector machine can predict the number of arginine methylation sites with 0.36 R -squared. In the third stage, experiments carried out with the updated arginine methylation site data set show that utilizing CTD features and adopting random forest as the classifier outperform previous methods. The accuracy of identification can reach 82.1 and 82.5% in single methylarginine and double methylarginine data sets, respectively. The discovery presented in this paper can be helpful for future research on arginine methylation.

PRMxAI: protein arginine methylation sites prediction based on amino acid spatial distribution using explainable artificial intelligence

“…The state-of-the-art predictors and the PRMxAI were assessed on the same data sets for an unbiased comparison. The result of the PRMxAI on the data sets utilized in [ 37 , 39 ] was estimated as shown in Tables 8 and 9 , respectively. For mono-methylarginine, the PRMxAI presented 87.17% accuracy, 5.07% higher than CTD-RF [ 37 ].…”

“…Kumar et al [ 36 ] proposed a prediction model named PRmePRed for arginine methylation based on structural and physicochemical properties using SVM. An arginine methylation prediction method, CTD-RF, developed by Hou et al [ 37 ] that integrates RF with distribution, composition, and transition features. Some of the researchers also used convolutional neural network (CNN) and long short-term memory (LSTM) deep learning algorithms for the prediction of arginine methylation sites [ 38 , 39 ].…”

MultiScale-CNN-4mCPred: a multi-scale CNN and adaptive embedding-based method for mouse genome DNA N4-methylcytosine prediction

“…The 5mC is formed by transferring a methyl group from S-adenyl methionine to the fifth carbon of a cytosine residue [ 13 ] and has been widely demonstrated to play an important role in the biological progression associated with diabetes, cancer, and some neurological diseases [ 14 – 16 ]. The well-known DNA 6mA is a process of adding a methyl group to the 6-th position of an adenine ring catalyzed by DNA methyltransferases [ 17 ]. The 6mA is an essential epigenetic modification.…”

“…The emergence of bioinformatics makes it possible to identify 4mC sites on a large scale. Hundreds of computational tools have been developed over the past thirty years for predicting post-translational modification [ 17 , 21 – 23 ], RNA modification [ 24 – 26 ], single-cell analysis [ 27 ], protein functions [ 28 , 29 ], as well as protein structure [ 30 ], gene selection [ 31 ], cancer diagnosis [ 32 ], and even food recommendation [ 33 ]. With the help of computers of powerful computational ability, the computational methods made great progress in the protein structure recognition, which is thought as one of best challenging tasks.…”

CNNArginineMe: A CNN structure for training models for predicting arginine methylation sites based on the One-Hot encoding of peptide sequence

“…The model MePred-RF integrated the random-forest algorithm with a sequence-based feature selection technique ( Wei et al, 2019 ). Hou and coworkers built a model to predict PRme sites based on composition-transition-distribution features ( Hou et al, 2020 ). In the deep-learning-based models, CapsNet contained a multi-layer CNN for predicting PRme sites, which outperformed other well-known tools in most cases ( Wang D. et al, 2019 ).…”