Show simple item record

dc.contributor.authorLi, Ran
dc.contributor.authorYu, Liangyong
dc.contributor.authorZhou, Bo
dc.contributor.authorZeng, Xiangrui
dc.contributor.authorWang, Zhenyu
dc.contributor.authorYang, Xiaoyan
dc.contributor.authorZhang, Jing
dc.contributor.authorGao, Xin
dc.contributor.authorJiang, Rui
dc.contributor.authorXu, Min
dc.date.accessioned2020-11-12T05:47:45Z
dc.date.available2020-11-12T05:47:45Z
dc.date.issued2020-11-11
dc.date.submitted2020-04-13
dc.identifier.citationLi, R., Yu, L., Zhou, B., Zeng, X., Wang, Z., Yang, X., … Xu, M. (2020). Few-shot learning for classification of novel macromolecular structures in cryo-electron tomograms. PLOS Computational Biology, 16(11), e1008227. doi:10.1371/journal.pcbi.1008227
dc.identifier.issn1553-7358
dc.identifier.doi10.1371/journal.pcbi.1008227
dc.identifier.urihttp://hdl.handle.net/10754/665918
dc.description.abstractCryo-electron tomography (cryo-ET) provides 3D visualization of subcellular components in the near-native state and at sub-molecular resolutions in single cells, demonstrating an increasingly important role in structural biology in situ. However, systematic recognition and recovery of macromolecular structures in cryo-ET data remain challenging as a result of low signal-to-noise ratio (SNR), small sizes of macromolecules, and high complexity of the cellular environment. Subtomogram structural classification is an essential step for such task. Although acquisition of large amounts of subtomograms is no longer an obstacle due to advances in automation of data collection, obtaining the same number of structural labels is both computation and labor intensive. On the other hand, existing deep learning based supervised classification approaches are highly demanding on labeled data and have limited ability to learn about new structures rapidly from data containing very few labels of such new structures. In this work, we propose a novel approach for subtomogram classification based on few-shot learning. With our approach, classification of unseen structures in the training data can be conducted given few labeled samples in test data through instance embedding. Experiments were performed on both simulated and real datasets. Our experimental results show that we can make inference on new structures given only five labeled samples for each class with a competitive accuracy (> 0.86 on the simulated dataset with SNR = 0.1), or even one sample with an accuracy of 0.7644. The results on real datasets are also promising with accuracy > 0.9 on both conditions and even up to 1 on one of the real datasets. Our approach achieves significant improvement compared with the baseline method and has strong capabilities of generalizing to other cellular components.
dc.description.sponsorshipThis work was supported in part by U.S. National Institutes of Health (NIH) grants P41GM103712 and R01GM134020, U.S. National Science Foundation (NSF) grants DBI-1949629 and IIS-2007595, and Mark Foundation 19-044-ASP. XZ was supported by a fellowship from Carnegie
dc.publisherPublic Library of Science (PLoS)
dc.relation.urlhttps://dx.plos.org/10.1371/journal.pcbi.1008227
dc.rightsThis is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.titleFew-shot learning for classification of novel macromolecular structures in cryo-electron tomograms
dc.typeArticle
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentStructural and Functional Bioinformatics Group
dc.identifier.journalPLOS Computational Biology
dc.eprint.versionPublisher's Version/PDF
dc.contributor.institutionDepartment of Automation, Tsinghua University, Beijing, China.
dc.contributor.institutionComputational Biology Department, Carnegie Mellon University, Pittsburgh, PA, USA.
dc.contributor.institutionDepartment of Biomedical Engineering, Yale University, New Haven, CT, USA.
dc.contributor.institutionDepartment of Computer Science, University of California Irvine, Irvine, CA, USA.
dc.identifier.volume16
dc.identifier.issue11
dc.identifier.pagese1008227
kaust.personGao, Xin
dc.date.accepted2020-08-08
dc.relation.issupplementedbygithub:xulabs/aitom
refterms.dateFOA2020-11-12T05:49:14Z
display.relations<b>Is Supplemented By:</b><br/> <ul><li><i>[Software]</i> <br/> Title: xulabs/aitom: AI for tomography. Publication Date: 2019-06-13. github: <a href="https://github.com/xulabs/aitom" >xulabs/aitom</a> Handle: <a href="http://hdl.handle.net/10754/667784" >10754/667784</a></a></li></ul>


Files in this item

Thumbnail
Name:
journal.pcbi.1008227.pdf
Size:
3.694Mb
Format:
PDF
Description:
Published version

This item appears in the following Collection(s)

Show simple item record

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Except where otherwise noted, this item's license is described as This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.