Folding Membrane Proteins by Deep Transfer Learning

Handle URI:
http://hdl.handle.net/10754/625815
Title:
Folding Membrane Proteins by Deep Transfer Learning
Authors:
Wang, Sheng; Li, Zhen; Yu, Yizhou; Xu, Jinbo
Abstract:
Computational elucidation of membrane protein (MP) structures is challenging partially due to lack of sufficient solved structures for homology modeling. Here, we describe a high-throughput deep transfer learning method that first predicts MP contacts by learning from non-MPs and then predicts 3D structure models using the predicted contacts as distance restraints. Tested on 510 non-redundant MPs, our method has contact prediction accuracy at least 0.18 better than existing methods, predicts correct folds for 218 MPs, and generates 3D models with root-mean-square deviation (RMSD) less than 4 and 5 Å for 57 and 108 MPs, respectively. A rigorous blind test in the continuous automated model evaluation project shows that our method predicted high-resolution 3D models for two recent test MPs of 210 residues with RMSD ∼2 Å. We estimated that our method could predict correct folds for 1,345–1,871 reviewed human multi-pass MPs including a few hundred new folds, which shall facilitate the discovery of drugs targeting at MPs.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computational Bioscience Research Center (CBRC)
Citation:
Wang S, Li Z, Yu Y, Xu J (2017) Folding Membrane Proteins by Deep Transfer Learning. Cell Systems 5: 202–211.e3. Available: http://dx.doi.org/10.1016/j.cels.2017.09.001.
Publisher:
Elsevier BV
Journal:
Cell Systems
Issue Date:
29-Aug-2017 ; 27-Sep-2017
DOI:
10.1101/181628; 10.1016/j.cels.2017.09.001
ARXIV:
arXiv:1708.08407
Type:
Article
ISSN:
2405-4712
Sponsors:
This work is supported by NIH grant R01GM089753 to J.X. and National Science Foundation grant DBI-1564955 to J.X. The authors are also grateful to the support of Nvidia and the computational resources provided by XSEDE. An early version of this paper was submitted to and peer reviewed at the 2017 Annual International Conference on Research in Computational Molecular Biology (RECOMB). The manuscript was revised and then independently further reviewed at Cell Systems.
Additional Links:
http://www.sciencedirect.com/science/article/pii/S2405471217303897
Appears in Collections:
Articles; Computational Bioscience Research Center (CBRC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, Shengen
dc.contributor.authorLi, Zhenen
dc.contributor.authorYu, Yizhouen
dc.contributor.authorXu, Jinboen
dc.date.accessioned2018-01-16T13:48:46Z-
dc.date.available2017-10-05T12:47:09Z-
dc.date.available2018-01-16T13:48:46Z-
dc.date.issued2017-08-29-
dc.date.issued2017-09-27en
dc.identifier.citationWang S, Li Z, Yu Y, Xu J (2017) Folding Membrane Proteins by Deep Transfer Learning. Cell Systems 5: 202–211.e3. Available: http://dx.doi.org/10.1016/j.cels.2017.09.001.en
dc.identifier.issn2405-4712en
dc.identifier.doi10.1101/181628-
dc.identifier.doi10.1016/j.cels.2017.09.001en
dc.identifier.urihttp://hdl.handle.net/10754/625815-
dc.description.abstractComputational elucidation of membrane protein (MP) structures is challenging partially due to lack of sufficient solved structures for homology modeling. Here, we describe a high-throughput deep transfer learning method that first predicts MP contacts by learning from non-MPs and then predicts 3D structure models using the predicted contacts as distance restraints. Tested on 510 non-redundant MPs, our method has contact prediction accuracy at least 0.18 better than existing methods, predicts correct folds for 218 MPs, and generates 3D models with root-mean-square deviation (RMSD) less than 4 and 5 Å for 57 and 108 MPs, respectively. A rigorous blind test in the continuous automated model evaluation project shows that our method predicted high-resolution 3D models for two recent test MPs of 210 residues with RMSD ∼2 Å. We estimated that our method could predict correct folds for 1,345–1,871 reviewed human multi-pass MPs including a few hundred new folds, which shall facilitate the discovery of drugs targeting at MPs.en
dc.description.sponsorshipThis work is supported by NIH grant R01GM089753 to J.X. and National Science Foundation grant DBI-1564955 to J.X. The authors are also grateful to the support of Nvidia and the computational resources provided by XSEDE. An early version of this paper was submitted to and peer reviewed at the 2017 Annual International Conference on Research in Computational Molecular Biology (RECOMB). The manuscript was revised and then independently further reviewed at Cell Systems.en
dc.language.isoenen
dc.publisherElsevier BVen
dc.relation.urlhttp://www.sciencedirect.com/science/article/pii/S2405471217303897en
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Cell Systems. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Cell Systems, [, , (2017-09-27)] DOI: 10.1016/j.cels.2017.09.001 . © 2017. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/en
dc.subjectmembrane protein foldingen
dc.subjectmembrane protein contact predictionen
dc.subjectdeep learningen
dc.subjectdeep transfer learningen
dc.subjecthomology modelingen
dc.subjectco-evolution analysisen
dc.subjectmultiple sequence alignmenten
dc.titleFolding Membrane Proteins by Deep Transfer Learningen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputational Bioscience Research Center (CBRC)en
dc.identifier.journalCell Systemsen
dc.eprint.versionPost-printen
dc.contributor.institutionDepartment of Human Genetics, University of Chicago, Chicago, IL 60637, USAen
dc.contributor.institutionToyota Technological Institute at Chicago, Chicago, IL 60637, USAen
dc.contributor.institutionDepartment of Computer Science, University of Hong Kong, Hong Kongen
dc.identifier.arxividarXiv:1708.08407-
kaust.authorWang, Shengen

Version History

VersionItem Editor Date Summary
2 10754/625815grenzdm2018-01-16 13:47:25.501Final version published in journal.
1 10754/625815.1grenzdm2017-10-05 12:47:09.0
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.