MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features
RNA modifications play crucial roles in various biological processes and diseases. Accurate prediction of RNA modification sites is essential for understanding their functions. In this study, we propose a hybrid approach that fuses a pre-trained sequence representation with various sequence features...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | RNA Biology |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/15476286.2024.2315384 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832539993719963648 |
|---|---|
| author | Linshu Wang Yuan Zhou |
| author_facet | Linshu Wang Yuan Zhou |
| author_sort | Linshu Wang |
| collection | DOAJ |
| description | RNA modifications play crucial roles in various biological processes and diseases. Accurate prediction of RNA modification sites is essential for understanding their functions. In this study, we propose a hybrid approach that fuses a pre-trained sequence representation with various sequence features to predict multiple types of RNA modifications in one combined prediction framework. We developed MRM-BERT, a deep learning method that combined the pre-trained DNABERT deep sequence representation module and the convolutional neural network (CNN) exploiting four traditional sequence feature encodings to improve the prediction performance. MRM-BERT was evaluated on multiple datasets of 12 commonly occurring RNA modifications, including m6A, m5C, m1A and so on. The results demonstrate that our hybrid model outperforms other models in terms of area under receiver operating characteristic curve (AUC) for all 12 types of RNA modifications. MRM-BERT is available as an online tool (http://117.122.208.21:8501) or source code (https://github.com/abhhba999/MRM-BERT), which allows users to predict RNA modification sites and visualize the results. Overall, our study provides an effective and efficient approach to predict multiple RNA modifications, contributing to the understanding of RNA biology and the development of therapeutic strategies. |
| format | Article |
| id | doaj-art-c3dd1a60abc44e19bf761e27f0ebdbb5 |
| institution | Kabale University |
| issn | 1547-6286 1555-8584 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | RNA Biology |
| spelling | doaj-art-c3dd1a60abc44e19bf761e27f0ebdbb52025-02-05T05:42:20ZengTaylor & Francis GroupRNA Biology1547-62861555-85842024-12-0121134034910.1080/15476286.2024.2315384MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence featuresLinshu Wang0Yuan Zhou1Department of Biomedical Informatics, School of Basic Medical Sciences, Peking University, Beijing, ChinaDepartment of Biomedical Informatics, School of Basic Medical Sciences, Peking University, Beijing, ChinaRNA modifications play crucial roles in various biological processes and diseases. Accurate prediction of RNA modification sites is essential for understanding their functions. In this study, we propose a hybrid approach that fuses a pre-trained sequence representation with various sequence features to predict multiple types of RNA modifications in one combined prediction framework. We developed MRM-BERT, a deep learning method that combined the pre-trained DNABERT deep sequence representation module and the convolutional neural network (CNN) exploiting four traditional sequence feature encodings to improve the prediction performance. MRM-BERT was evaluated on multiple datasets of 12 commonly occurring RNA modifications, including m6A, m5C, m1A and so on. The results demonstrate that our hybrid model outperforms other models in terms of area under receiver operating characteristic curve (AUC) for all 12 types of RNA modifications. MRM-BERT is available as an online tool (http://117.122.208.21:8501) or source code (https://github.com/abhhba999/MRM-BERT), which allows users to predict RNA modification sites and visualize the results. Overall, our study provides an effective and efficient approach to predict multiple RNA modifications, contributing to the understanding of RNA biology and the development of therapeutic strategies.https://www.tandfonline.com/doi/10.1080/15476286.2024.2315384RNA modificationsdeep learningsequence featuresBidirectional encoder representations from transformers (BERT)convolutional neural networkweb server |
| spellingShingle | Linshu Wang Yuan Zhou MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features RNA Biology RNA modifications deep learning sequence features Bidirectional encoder representations from transformers (BERT) convolutional neural network web server |
| title | MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features |
| title_full | MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features |
| title_fullStr | MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features |
| title_full_unstemmed | MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features |
| title_short | MRM-BERT: a novel deep neural network predictor of multiple RNA modifications by fusing BERT representation and sequence features |
| title_sort | mrm bert a novel deep neural network predictor of multiple rna modifications by fusing bert representation and sequence features |
| topic | RNA modifications deep learning sequence features Bidirectional encoder representations from transformers (BERT) convolutional neural network web server |
| url | https://www.tandfonline.com/doi/10.1080/15476286.2024.2315384 |
| work_keys_str_mv | AT linshuwang mrmbertanoveldeepneuralnetworkpredictorofmultiplernamodificationsbyfusingbertrepresentationandsequencefeatures AT yuanzhou mrmbertanoveldeepneuralnetworkpredictorofmultiplernamodificationsbyfusingbertrepresentationandsequencefeatures |