BiLSTM- and CNN-Based m6A Modification Prediction Model for circRNAs

Yuan,Tang,Li,Lang,Song,Yang,Zhou

DOI: https://doi.org/10.3390/molecules29112429

IF: 4.6

2024-05-22

Molecules

Abstract:m6A methylation, a ubiquitous modification on circRNAs, exerts a profound influence on RNA function, intracellular behavior, and diverse biological processes, including disease development. While prediction algorithms exist for mRNA m6A modifications, a critical gap remains in the prediction of circRNA m6A modifications. Therefore, accurate identification and prediction of m6A sites are imperative for understanding RNA function and regulation. This study presents a novel hybrid model combining a convolutional neural network (CNN) and a bidirectional long short-term memory network (BiLSTM) for precise m6A methylation site prediction in circular RNAs (circRNAs) based on data from HEK293 cells. This model exploits the synergy between CNN's ability to extract intricate sequence features and BiLSTM's strength in capturing long-range dependencies. Furthermore, the integrated attention mechanism empowers the model to pinpoint critical biological information for studying circRNA m6A methylation. Our model, exhibiting over 78% prediction accuracy on independent datasets, offers not only a valuable tool for scientific research but also a strong foundation for future biomedical applications. This work not only furthers our understanding of gene expression regulation but also opens new avenues for the exploration of circRNA methylation in biological research.

chemistry, multidisciplinary,biochemistry & molecular biology

What problem does this paper attempt to address?

The paper aims to address the following key issues: ### Research Background and Objectives - **Research Background**: N6-methyladenosine (m6A), a modification commonly found on circular RNA (circRNA), plays a significant role in RNA function, intracellular behavior, and various biological processes, including disease development. Although there are existing prediction algorithms for m6A modifications on mRNA, there is a noticeable gap in the prediction of m6A modifications on circRNA. - **Research Objectives**: To propose a hybrid model combining Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory Network (BiLSTM) for accurately predicting m6A methylation sites in circular RNA (circRNA). This model aims to fill the gap in the field of circRNA m6A modification prediction and significantly improve prediction accuracy. ### Main Contributions - **Methodological Innovation**: This study proposes a novel hybrid model that utilizes CNN to extract complex sequence features and BiLSTM to capture long-range dependencies. Additionally, an attention mechanism is introduced to identify key biological information. - **Experimental Validation**: The model demonstrates over 78% prediction accuracy on independent datasets, providing a valuable tool for scientific research and laying a solid foundation for future biomedical applications. - **Biological Significance**: This work not only advances our understanding of gene expression regulation but also opens new avenues for exploring circRNA methylation in biological research. ### Summary This paper attempts to fill the gap in the field of circRNA m6A modification prediction by developing a novel prediction model. The model combines the advantages of CNN and BiLSTM and incorporates an attention mechanism, enabling efficient and accurate prediction of m6A modification sites in circular RNA. Empirical studies have demonstrated the model's effectiveness and reliability, which is of great significance for understanding the function of circRNA and its role in biological processes.

BiLSTM- and CNN-Based m6A Modification Prediction Model for circRNAs

A CNN-Based RNA N6-Methyladenosine Site Predictor for Multiple Species Using Heterogeneous Features Representation

Advances in detecting N6-methyladenosine modification in circRNAs

Deepm6A-MT: A deep learning-based method for identifying RNA N6-methyladenosine sites in multiple tissues

A Combined Deep Learning Framework for Mammalian M6a Site Prediction

m6Aminer: Predicting the m6Am Sites on mRNA by Fusing Multiple Sequence-Derived Features into a CatBoost-Based Classifier

EDLm6APred: ensemble deep learning approach for mRNA m6A site prediction

m5UMCB: Prediction of RNA 5-methyluridine sites using multi-scale convolutional neural network with BiLSTM

Rm-LR: A long-range-based deep learning model for predicting multiple types of RNA modifications

EMDL_m6Am: identifying N6,2'-O-dimethyladenosine sites based on stacking ensemble deep learning

Prediction of N6-methyladenosine sites using convolution neural network model based on distributed feature representations

MTTLm6A: A multi-task transfer learning approach for base-resolution mRNA m6A site prediction based on an improved transformer

CircSSNN: circRNA-binding site prediction via sequence self-attention neural networks with pre-normalization

MSCAN: multi-scale self- and cross-attention network for RNA methylation site prediction

An Integrated TCN-CrossMHA Model for Predicting circRNA-RBP Binding Sites

Comprehensive Review and Assessment of Computational Methods for Prediction of N6-Methyladenosine Sites

BJLD-CMI: a predictive circRNA-miRNA interactions model combining multi-angle feature information

DLm6Am: A Deep-Learning-Based Tool for Identifying N6,2'-O-Dimethyladenosine Sites in RNA Sequences

Novel insights into the interaction between N6-methyladenosine modification and circular RNA

CRIECNN: Ensemble convolutional neural network and advanced feature extraction methods for the precise forecasting of circRNA-RBP binding sites

A novel circRNA-miRNA association prediction model based on structural deep neural network embedding