Toward Deep Learning Approaches for Learning Structure Motifs and Classifying Biological Sequences From RNA A-to-I Editing Events

Abstract

RNA editing is a post-transcriptional RNA sequence modification that alters the mature RNA sequence from its template DNA sequences. RNA editing events are critical in various biological and biochemical mechanisms, and can expand the transcriptomic and proteomic diversity from altered gene regulation to mutations. A-to-I RNA editing is now being vastly detected and quantified on a global scale and gained much attention. A deeper understanding of this process with insufficient genomic annotations and prior knowledge-based filtering steps, such as high-throughput next-generation sequencing techniques, are needed, in addition to data regarding whether an editing location is located in one of the following three main classes of RNA editing sites: ALU in Alu repetitive elements, REP in non-Alu repetitive elements, and NONREP in non-repetitive regions. This study proposes deep learning approaches to ameliorate these issues by learning motif patterns and identifying regions of A-to-I editing events in biological sequences. Using datasets derived from the public REDI portal, 300,000 editing sites were equally divided and used for learning sequence and structure motifs visualized by convolutional kernels. We explored the RNA editing pattern changes using information of positional and class enrichment of learned motifs in the three aforementioned classes. We demonstrated that these newly investigated approaches using large-scale RNA sequencing data offer excellent classification accuracy with a well-optimized convolutional neural network and recurrent neural network classifiers that obtained average area under curves of 0.960 and 0.962, respectively. The findings further decipher the principles underlying RNA editing events and will facilitate more effective RNA sequencing research.

Heuristic Analysis of Genomic Sequence Processing Models for High Efficiency Prediction: A Statistical Perspective

“…PLS with novel TTZ feature vector [ 51 ] is applied to lung disease classification and is able to achieve an accuracy of 92.65%, while the LASSO SVM model is applied to cancer datasets and is able to achieve an accuracy of 91.3%, thereby making them useful for various applications. Similar models have been explained, where the authors have discussed the use of repeated incremental pruning to produce error reduction (RIPPER) [ 52 ], CNNs, RNNs [ 53 ], and RIPPER with SVM for multiple applications [ 54 ]. The RIPPER model is capable of achieving an accuracy of 80.8%, CNNs an accuracy of 96%, RNNs an accuracy of 96.2%, and SVM with RIPPER to achieve an accuracy of 99.7%, thereby improving their utility for real-time clinical and on-field deployments.…”

“…( 5 ) is presented, wherein the accuracy of different genome models for heart and brain sequence data analysis is visualized. It is observed that RIPPER SVM [ 54 ], SHoT [ 27 ], RNN [ 53 ], and CNN [ 53 ] outperform other models in terms of accuracy, and thus must be used for heart and brain dataset genome sequence analysis.…”

“…It can be observed from Fig. ( 6 ) that RNN [ 53 ], CNN [ 53 ], GA SIWR [ 2 ], and SpineNet-6m [ 5 ] outperform others for grain genome dataset. A similar representation for genome analysis of non-grains can be observed (Fig.…”