MSCAN: multi-scale self- and cross-attention network for RNA methylation site prediction
- PMID: 38233745
- PMCID: PMC10795237
- DOI: 10.1186/s12859-024-05649-1
MSCAN: multi-scale self- and cross-attention network for RNA methylation site prediction
Abstract
Background: Epi-transcriptome regulation through post-transcriptional RNA modifications is essential for all RNA types. Precise recognition of RNA modifications is critical for understanding their functions and regulatory mechanisms. However, wet experimental methods are often costly and time-consuming, limiting their wide range of applications. Therefore, recent research has focused on developing computational methods, particularly deep learning (DL). Bidirectional long short-term memory (BiLSTM), convolutional neural network (CNN), and the transformer have demonstrated achievements in modification site prediction. However, BiLSTM cannot achieve parallel computation, leading to a long training time, CNN cannot learn the dependencies of the long distance of the sequence, and the Transformer lacks information interaction with sequences at different scales. This insight underscores the necessity for continued research and development in natural language processing (NLP) and DL to devise an enhanced prediction framework that can effectively address the challenges presented.
Results: This study presents a multi-scale self- and cross-attention network (MSCAN) to identify the RNA methylation site using an NLP and DL way. Experiment results on twelve RNA modification sites (m6A, m1A, m5C, m5U, m6Am, m7G, Ψ, I, Am, Cm, Gm, and Um) reveal that the area under the receiver operating characteristic of MSCAN obtains respectively 98.34%, 85.41%, 97.29%, 96.74%, 99.04%, 79.94%, 76.22%, 65.69%, 92.92%, 92.03%, 95.77%, 89.66%, which is better than the state-of-the-art prediction model. This indicates that the model has strong generalization capabilities. Furthermore, MSCAN reveals a strong association among different types of RNA modifications from an experimental perspective. A user-friendly web server for predicting twelve widely occurring human RNA modification sites (m6A, m1A, m5C, m5U, m6Am, m7G, Ψ, I, Am, Cm, Gm, and Um) is available at http://47.242.23.141/MSCAN/index.php .
Conclusions: A predictor framework has been developed through binary classification to predict RNA methylation sites.
Keywords: Cross-attention; Multi-scale; Predictor; RNA methylation; Self-attention; Transformer.
© 2024. The Author(s).
Conflict of interest statement
The authors declare that they have no competing interests.
Figures
![Fig. 1](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig1_HTML.gif)
![Fig. 2](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig2_HTML.gif)
![Fig. 3](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig3_HTML.gif)
![Fig. 4](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig4_HTML.gif)
![Fig. 5](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig5_HTML.gif)
![Fig. 6](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig6_HTML.gif)
![Fig. 7](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig7a_HTML.gif)
![Fig. 7](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig7a_HTML.gif)
![Fig. 8](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig8_HTML.gif)
![Fig. 9](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig9_HTML.gif)
![Fig. 10](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/10795237/bin/12859_2024_5649_Fig10_HTML.gif)
Similar articles
-
MTTLm6A: A multi-task transfer learning approach for base-resolution mRNA m6A site prediction based on an improved transformer.Math Biosci Eng. 2024 Jan;21(1):272-299. doi: 10.3934/mbe.2024013. Epub 2022 Dec 12. Math Biosci Eng. 2024. PMID: 38303423
-
Mini-review: Recent advances in post-translational modification site prediction based on deep learning.Comput Struct Biotechnol J. 2022 Jun 30;20:3522-3532. doi: 10.1016/j.csbj.2022.06.045. eCollection 2022. Comput Struct Biotechnol J. 2022. PMID: 35860402 Free PMC article. Review.
-
EMDLP: Ensemble multiscale deep learning model for RNA methylation site prediction.BMC Bioinformatics. 2022 Jun 8;23(1):221. doi: 10.1186/s12859-022-04756-1. BMC Bioinformatics. 2022. PMID: 35676633 Free PMC article.
-
Attention-based multi-label neural networks for integrated prediction and interpretation of twelve widely occurring RNA modifications.Nat Commun. 2021 Jun 29;12(1):4011. doi: 10.1038/s41467-021-24313-3. Nat Commun. 2021. PMID: 34188054 Free PMC article.
-
Comprehensive review and assessment of computational methods for predicting RNA post-transcriptional modification sites from RNA sequences.Brief Bioinform. 2020 Sep 25;21(5):1676-1696. doi: 10.1093/bib/bbz112. Brief Bioinform. 2020. PMID: 31714956 Review.
Cited by
-
Interpretable Multi-Scale Deep Learning for RNA Methylation Analysis across Multiple Species.Int J Mol Sci. 2024 Mar 1;25(5):2869. doi: 10.3390/ijms25052869. Int J Mol Sci. 2024. PMID: 38474116 Free PMC article.
References
-
- Wang H, Wang SY, Zhang Y, Bi SD, Zhu XL. A brief review of machine learning methods for RNA methylation sites prediction. Methods. 2022;203:399–421. - PubMed
-
- Chen LF, Tan XQ, Wang DY, Zhong FS, Liu XH, Yang TB, Luo XM, Chen KX, Jiang HL, Zheng MY. TransformerCPI: improving compound–protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments. Bioinformatics. 2020;36(16):4406–4414. - PubMed
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources