MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain
| Publication Date: |
2018-03-06
|
|---|---|
| Publisher: |
Oxford University Press
|
| Print ISSN: |
1367-4803
|
| Electronic ISSN: |
1460-2059
|
| Topics: |
Biology
Computer Science
Medicine
|
| Published by: |
| _version_ | 1836398817050624000 |
|---|---|
| autor | Yang J, Shen H. |
| beschreibung | Motivation Inter-residue contacts in proteins have been widely acknowledged to be valuable for protein 3 D structure prediction. Accurate prediction of long-range transmembrane inter-helix residue contacts can significantly improve the quality of simulated membrane protein models. Results In this paper, we present an updated MemBrain predictor, which aims to predict transmembrane protein residue contacts. Our new model benefits from an efficient learning algorithm that can mine latent structural features, which exist in original feature space. The new MemBrain is a two-stage inter-helix contact predictor. The first stage takes sequence-based features as inputs and outputs coarse contact probabilities for each residue pair, which will be further fed into convolutional neural network together with predictions from three direct-coupling analysis approaches in the second stage. Experimental results on the training dataset show that our method achieves an average accuracy of 81.6% for the top L /5 predictions using a strict sequence-based jackknife cross-validation. Evaluated on the test dataset, MemBrain can achieve 79.4% prediction accuracy. Moreover, for the top L /5 predicted long-range loop contacts, the prediction performance can reach an accuracy of 56.4%. These results demonstrate that the new MemBrain is promising for transmembrane protein’s contact map prediction. Availability and implementation http://www.csbio.sjtu.edu.cn/bioinf/MemBrain/ Contact hbshen@sjtu.edu.cn Supplementary information Supplementary dataSupplementary data are available at Bioinformatics online. |
| citation_standardnr | 6180430 |
| datenlieferant | ipn_articles |
| feed_id | 2184 |
| feed_publisher | Oxford University Press |
| feed_publisher_url | http://global.oup.com/ |
| insertion_date | 2018-03-06 |
| journaleissn | 1460-2059 |
| journalissn | 1367-4803 |
| publikationsjahr_anzeige | 2018 |
| publikationsjahr_facette | 2018 |
| publikationsjahr_intervall | 7984:2015-2019 |
| publikationsjahr_sort | 2018 |
| publisher | Oxford University Press |
| quelle | Bioinformatics |
| relation | https://academic.oup.com/bioinformatics/article/34/2/230/4158791?rss=1 |
| search_space | articles |
| shingle_author_1 | Yang J, Shen H. |
| shingle_author_2 | Yang J, Shen H. |
| shingle_author_3 | Yang J, Shen H. |
| shingle_author_4 | Yang J, Shen H. |
| shingle_catch_all_1 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain Motivation Inter-residue contacts in proteins have been widely acknowledged to be valuable for protein 3 D structure prediction. Accurate prediction of long-range transmembrane inter-helix residue contacts can significantly improve the quality of simulated membrane protein models. Results In this paper, we present an updated MemBrain predictor, which aims to predict transmembrane protein residue contacts. Our new model benefits from an efficient learning algorithm that can mine latent structural features, which exist in original feature space. The new MemBrain is a two-stage inter-helix contact predictor. The first stage takes sequence-based features as inputs and outputs coarse contact probabilities for each residue pair, which will be further fed into convolutional neural network together with predictions from three direct-coupling analysis approaches in the second stage. Experimental results on the training dataset show that our method achieves an average accuracy of 81.6% for the top L /5 predictions using a strict sequence-based jackknife cross-validation. Evaluated on the test dataset, MemBrain can achieve 79.4% prediction accuracy. Moreover, for the top L /5 predicted long-range loop contacts, the prediction performance can reach an accuracy of 56.4%. These results demonstrate that the new MemBrain is promising for transmembrane protein’s contact map prediction. Availability and implementation http://www.csbio.sjtu.edu.cn/bioinf/MemBrain/ Contact hbshen@sjtu.edu.cn Supplementary information Supplementary dataSupplementary data are available at Bioinformatics online. Yang J, Shen H. Oxford University Press 1367-4803 13674803 1460-2059 14602059 |
| shingle_catch_all_2 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain Motivation Inter-residue contacts in proteins have been widely acknowledged to be valuable for protein 3 D structure prediction. Accurate prediction of long-range transmembrane inter-helix residue contacts can significantly improve the quality of simulated membrane protein models. Results In this paper, we present an updated MemBrain predictor, which aims to predict transmembrane protein residue contacts. Our new model benefits from an efficient learning algorithm that can mine latent structural features, which exist in original feature space. The new MemBrain is a two-stage inter-helix contact predictor. The first stage takes sequence-based features as inputs and outputs coarse contact probabilities for each residue pair, which will be further fed into convolutional neural network together with predictions from three direct-coupling analysis approaches in the second stage. Experimental results on the training dataset show that our method achieves an average accuracy of 81.6% for the top L /5 predictions using a strict sequence-based jackknife cross-validation. Evaluated on the test dataset, MemBrain can achieve 79.4% prediction accuracy. Moreover, for the top L /5 predicted long-range loop contacts, the prediction performance can reach an accuracy of 56.4%. These results demonstrate that the new MemBrain is promising for transmembrane protein’s contact map prediction. Availability and implementation http://www.csbio.sjtu.edu.cn/bioinf/MemBrain/ Contact hbshen@sjtu.edu.cn Supplementary information Supplementary dataSupplementary data are available at Bioinformatics online. Yang J, Shen H. Oxford University Press 1367-4803 13674803 1460-2059 14602059 |
| shingle_catch_all_3 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain Motivation Inter-residue contacts in proteins have been widely acknowledged to be valuable for protein 3 D structure prediction. Accurate prediction of long-range transmembrane inter-helix residue contacts can significantly improve the quality of simulated membrane protein models. Results In this paper, we present an updated MemBrain predictor, which aims to predict transmembrane protein residue contacts. Our new model benefits from an efficient learning algorithm that can mine latent structural features, which exist in original feature space. The new MemBrain is a two-stage inter-helix contact predictor. The first stage takes sequence-based features as inputs and outputs coarse contact probabilities for each residue pair, which will be further fed into convolutional neural network together with predictions from three direct-coupling analysis approaches in the second stage. Experimental results on the training dataset show that our method achieves an average accuracy of 81.6% for the top L /5 predictions using a strict sequence-based jackknife cross-validation. Evaluated on the test dataset, MemBrain can achieve 79.4% prediction accuracy. Moreover, for the top L /5 predicted long-range loop contacts, the prediction performance can reach an accuracy of 56.4%. These results demonstrate that the new MemBrain is promising for transmembrane protein’s contact map prediction. Availability and implementation http://www.csbio.sjtu.edu.cn/bioinf/MemBrain/ Contact hbshen@sjtu.edu.cn Supplementary information Supplementary dataSupplementary data are available at Bioinformatics online. Yang J, Shen H. Oxford University Press 1367-4803 13674803 1460-2059 14602059 |
| shingle_catch_all_4 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain Motivation Inter-residue contacts in proteins have been widely acknowledged to be valuable for protein 3 D structure prediction. Accurate prediction of long-range transmembrane inter-helix residue contacts can significantly improve the quality of simulated membrane protein models. Results In this paper, we present an updated MemBrain predictor, which aims to predict transmembrane protein residue contacts. Our new model benefits from an efficient learning algorithm that can mine latent structural features, which exist in original feature space. The new MemBrain is a two-stage inter-helix contact predictor. The first stage takes sequence-based features as inputs and outputs coarse contact probabilities for each residue pair, which will be further fed into convolutional neural network together with predictions from three direct-coupling analysis approaches in the second stage. Experimental results on the training dataset show that our method achieves an average accuracy of 81.6% for the top L /5 predictions using a strict sequence-based jackknife cross-validation. Evaluated on the test dataset, MemBrain can achieve 79.4% prediction accuracy. Moreover, for the top L /5 predicted long-range loop contacts, the prediction performance can reach an accuracy of 56.4%. These results demonstrate that the new MemBrain is promising for transmembrane protein’s contact map prediction. Availability and implementation http://www.csbio.sjtu.edu.cn/bioinf/MemBrain/ Contact hbshen@sjtu.edu.cn Supplementary information Supplementary dataSupplementary data are available at Bioinformatics online. Yang J, Shen H. Oxford University Press 1367-4803 13674803 1460-2059 14602059 |
| shingle_title_1 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| shingle_title_2 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| shingle_title_3 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| shingle_title_4 | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| timestamp | 2025-06-30T23:33:06.326Z |
| titel | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| titel_suche | MemBrain-contact 2.0: a new two-stage machine learning model for the prediction enhancement of transmembrane protein residue contacts in the full chain |
| topic | W SQ-SU WW-YZ |
| uid | ipn_articles_6180430 |