Research on RNA secondary structure predicting via bidirectional recurrent neural network
Abstract Background RNA secondary structure prediction is an important research content in the field of biological information. Predicting RNA secondary structure with pseudoknots has been proved to be an NP-hard problem. Traditional machine learning methods can not effectively apply protein sequenc...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2021-09-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-021-04332-z |
| _version_ | 1818364137380511744 |
|---|---|
| author | Weizhong Lu Yan Cao Hongjie Wu Yijie Ding Zhengwei Song Yu Zhang Qiming Fu Haiou Li |
| author_facet | Weizhong Lu Yan Cao Hongjie Wu Yijie Ding Zhengwei Song Yu Zhang Qiming Fu Haiou Li |
| author_sort | Weizhong Lu |
| collection | DOAJ |
| description | Abstract Background RNA secondary structure prediction is an important research content in the field of biological information. Predicting RNA secondary structure with pseudoknots has been proved to be an NP-hard problem. Traditional machine learning methods can not effectively apply protein sequence information with different sequence lengths to the prediction process due to the constraint of the self model when predicting the RNA secondary structure. In addition, there is a large difference between the number of paired bases and the number of unpaired bases in the RNA sequences, which means the problem of positive and negative sample imbalance is easy to make the model fall into a local optimum. To solve the above problems, this paper proposes a variable-length dynamic bidirectional Gated Recurrent Unit(VLDB GRU) model. The model can accept sequences with different lengths through the introduction of flag vector. The model can also make full use of the base information before and after the predicted base and can avoid losing part of the information due to truncation. Introducing a weight vector to predict the RNA training set by dynamically adjusting each base loss function solves the problem of balanced sample imbalance. Results The algorithm proposed in this paper is compared with the existing algorithms on five representative subsets of the data set RNA STRAND. The experimental results show that the accuracy and Matthews correlation coefficient of the method are improved by 4.7% and 11.4%, respectively. Conclusions The flag vector introduced allows the model to effectively use the information before and after the protein sequence; the introduced weight vector solves the problem of unbalanced sample balance. Compared with other algorithms, the LVDB GRU algorithm proposed in this paper has the best detection results. |
| first_indexed | 2024-12-13T21:59:35Z |
| format | Article |
| id | doaj.art-4143362213044d99b78244c1bb873eda |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-12-13T21:59:35Z |
| publishDate | 2021-09-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-4143362213044d99b78244c1bb873eda2022-12-21T23:30:03ZengBMCBMC Bioinformatics1471-21052021-09-0122S311810.1186/s12859-021-04332-zResearch on RNA secondary structure predicting via bidirectional recurrent neural networkWeizhong Lu0Yan Cao1Hongjie Wu2Yijie Ding3Zhengwei Song4Yu Zhang5Qiming Fu6Haiou Li7School of Electronic and Information Engineering, Suzhou University of Science and TechnologySchool of Electronic and Information Engineering, Suzhou University of Science and TechnologySchool of Electronic and Information Engineering, Suzhou University of Science and TechnologySchool of Electronic and Information Engineering, Suzhou University of Science and TechnologySchool of Electronic and Information Engineering, Suzhou University of Science and TechnologySuzhou Industrial Park Institute of Services OutsourcingSchool of Electronic and Information Engineering, Suzhou University of Science and TechnologySchool of Electronic and Information Engineering, Suzhou University of Science and TechnologyAbstract Background RNA secondary structure prediction is an important research content in the field of biological information. Predicting RNA secondary structure with pseudoknots has been proved to be an NP-hard problem. Traditional machine learning methods can not effectively apply protein sequence information with different sequence lengths to the prediction process due to the constraint of the self model when predicting the RNA secondary structure. In addition, there is a large difference between the number of paired bases and the number of unpaired bases in the RNA sequences, which means the problem of positive and negative sample imbalance is easy to make the model fall into a local optimum. To solve the above problems, this paper proposes a variable-length dynamic bidirectional Gated Recurrent Unit(VLDB GRU) model. The model can accept sequences with different lengths through the introduction of flag vector. The model can also make full use of the base information before and after the predicted base and can avoid losing part of the information due to truncation. Introducing a weight vector to predict the RNA training set by dynamically adjusting each base loss function solves the problem of balanced sample imbalance. Results The algorithm proposed in this paper is compared with the existing algorithms on five representative subsets of the data set RNA STRAND. The experimental results show that the accuracy and Matthews correlation coefficient of the method are improved by 4.7% and 11.4%, respectively. Conclusions The flag vector introduced allows the model to effectively use the information before and after the protein sequence; the introduced weight vector solves the problem of unbalanced sample balance. Compared with other algorithms, the LVDB GRU algorithm proposed in this paper has the best detection results.https://doi.org/10.1186/s12859-021-04332-zRecurrent neural networkRNA secondary structure predictionPseudoknots |
| spellingShingle | Weizhong Lu Yan Cao Hongjie Wu Yijie Ding Zhengwei Song Yu Zhang Qiming Fu Haiou Li Research on RNA secondary structure predicting via bidirectional recurrent neural network BMC Bioinformatics Recurrent neural network RNA secondary structure prediction Pseudoknots |
| title | Research on RNA secondary structure predicting via bidirectional recurrent neural network |
| title_full | Research on RNA secondary structure predicting via bidirectional recurrent neural network |
| title_fullStr | Research on RNA secondary structure predicting via bidirectional recurrent neural network |
| title_full_unstemmed | Research on RNA secondary structure predicting via bidirectional recurrent neural network |
| title_short | Research on RNA secondary structure predicting via bidirectional recurrent neural network |
| title_sort | research on rna secondary structure predicting via bidirectional recurrent neural network |
| topic | Recurrent neural network RNA secondary structure prediction Pseudoknots |
| url | https://doi.org/10.1186/s12859-021-04332-z |
| work_keys_str_mv | AT weizhonglu researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT yancao researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT hongjiewu researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT yijieding researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT zhengweisong researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT yuzhang researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT qimingfu researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork AT haiouli researchonrnasecondarystructurepredictingviabidirectionalrecurrentneuralnetwork |