{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,9]],"date-time":"2025-11-09T07:50:01Z","timestamp":1762674601295,"version":"3.37.3"},"reference-count":20,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T00:00:00Z","timestamp":1661817600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T00:00:00Z","timestamp":1661817600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100002347","name":"Bundesministerium f\u00fcr Bildung und Forschung","doi-asserted-by":"publisher","award":["3FO18501"],"award-info":[{"award-number":["3FO18501"]}],"id":[{"id":"10.13039\/501100002347","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Berlin Institute for the Foundations of Learning and Data","award":["BIFOLD-BZML 01IS18037H"],"award-info":[{"award-number":["BIFOLD-BZML 01IS18037H"]}]},{"name":"Zuse-Institut Berlin"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"abstract":"Abstract<\/jats:title>\n Background<\/jats:title>\n Despite recent advances in cellular cryo-electron tomography (CET), developing automated tools for macromolecule identification in submolecular resolution remains challenging due to the lack of annotated data and high structural complexities. To date, the extent of the deep learning methods constructed for this problem is limited to conventional Convolutional Neural Networks (CNNs). Identifying macromolecules of different types and sizes is a tedious and time-consuming task. In this paper, we employ a capsule-based architecture to automate the task of macromolecule identification, that we refer to as 3D-UCaps. In particular, the architecture is composed of three components: feature extractor, capsule encoder, and CNN decoder. The feature extractor converts voxel intensities of input sub-tomograms to activities of local features. The encoder is a 3D Capsule Network (CapsNet) that takes local features to generate a low-dimensional representation of the input. Then, a 3D CNN decoder reconstructs the sub-tomograms from the given representation by upsampling.<\/jats:p>\n <\/jats:sec>\n Results<\/jats:title>\n We performed binary and multi-class localization and identification tasks on synthetic and experimental data. We observed that the 3D-UNet and the 3D-UCaps had an $$F_1-$$<\/jats:tex-math>\n \n \n F<\/mml:mi>\n 1<\/mml:mn>\n <\/mml:msub>\n -<\/mml:mo>\n <\/mml:mrow>\n <\/mml:math><\/jats:alternatives><\/jats:inline-formula>score mostly above 60% and 70%, respectively, on the test data. In both network architectures, we observed degradation of at least 40% in the $$F_1$$<\/jats:tex-math>\n \n F<\/mml:mi>\n 1<\/mml:mn>\n <\/mml:msub>\n <\/mml:math><\/jats:alternatives><\/jats:inline-formula>-score when identifying very small particles (PDB entry 3GL1) compared to a large particle (PDB entry 4D8Q). In the multi-class identification task of experimental data, 3D-UCaps had an $$F_1$$<\/jats:tex-math>\n \n F<\/mml:mi>\n 1<\/mml:mn>\n <\/mml:msub>\n <\/mml:math><\/jats:alternatives><\/jats:inline-formula>-score of 91% on the test data in contrast to 64% of the 3D-UNet. The better $$F_1$$<\/jats:tex-math>\n \n F<\/mml:mi>\n 1<\/mml:mn>\n <\/mml:msub>\n <\/mml:math><\/jats:alternatives><\/jats:inline-formula>-score of 3D-UCaps compared to 3D-UNet is obtained by a higher precision score. We speculate this to be due to the capsule network employed in the encoder. To study the effect of the CapsNet-based encoder architecture further, we performed an ablation study and perceived that the $$F_1$$<\/jats:tex-math>\n \n F<\/mml:mi>\n 1<\/mml:mn>\n <\/mml:msub>\n <\/mml:math><\/jats:alternatives><\/jats:inline-formula>-score is boosted as network depth is increased which is in contrast to the previously reported results for the 3D-UNet. To present a reproducible work, source code, trained models, data as well as visualization results are made publicly available.<\/jats:p>\n <\/jats:sec>\n Conclusion<\/jats:title>\n Quantitative and qualitative results show that 3D-UCaps successfully perform various downstream tasks including identification and localization of macromolecules and can at least compete with CNN architectures for this task. Given that the capsule layers extract both the existence probability and the orientation of the molecules, this architecture has the potential to lead to representations of the data that are better interpretable than those of 3D-UNet.<\/jats:p>\n <\/jats:sec>","DOI":"10.1186\/s12859-022-04901-w","type":"journal-article","created":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T21:06:26Z","timestamp":1661893586000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Volumetric macromolecule identification in cryo-electron tomograms using capsule networks"],"prefix":"10.1186","volume":"23","author":[{"given":"Noushin","family":"Hajarolasvadi","sequence":"first","affiliation":[]},{"given":"Vikram","family":"Sunkara","sequence":"additional","affiliation":[]},{"given":"Sagar","family":"Khavnekar","sequence":"additional","affiliation":[]},{"given":"Florian","family":"Beck","sequence":"additional","affiliation":[]},{"given":"Robert","family":"Brandt","sequence":"additional","affiliation":[]},{"given":"Daniel","family":"Baum","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2022,8,30]]},"reference":[{"issue":"8","key":"4901_CR1","doi-asserted-by":"publisher","first-page":"1227","DOI":"10.1007\/s00138-018-0949-4","volume":"29","author":"C Che","year":"2018","unstructured":"Che C, Lin R, Zeng X, Elmaaroufi K, Galeotti J, Xu M. Improved deep learning-based macromolecules structure classification from electron cryo-tomograms. Mach Vis Appl. 2018;29(8):1227\u201336.","journal-title":"Mach Vis Appl"},{"issue":"11","key":"4901_CR2","doi-asserted-by":"publisher","first-page":"1386","DOI":"10.1038\/s41592-021-01275-4","volume":"18","author":"E Moebel","year":"2021","unstructured":"Moebel E, Martinez-Sanchez A, Lamm L, Righetto RD, Wietrzynski W, Albert S, Larivi\u00e8re D, Fourmentin E, Pfeffer S, Ortiz J, et al. Deep learning improves macromolecule identification in 3d cellular cryo-electron tomograms. Nat Methods. 2021;18(11):1386\u201394.","journal-title":"Nat Methods"},{"issue":"1","key":"4901_CR3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12859-020-03809-7","volume":"21","author":"A Al-Azzawi","year":"2020","unstructured":"Al-Azzawi A, Ouadou A, Max H, Duan Y, Tanner JJ, Cheng J. DeepCryoPicker: fully automated deep neural network for single protein particle picking in cryo-EM. BMC Bioinf. 2020;21(1):1\u201338.","journal-title":"BMC Bioinf"},{"key":"4901_CR4","doi-asserted-by":"publisher","first-page":"615","DOI":"10.1016\/S0091-679X(06)79025-2","volume":"79","author":"C Best","year":"2007","unstructured":"Best C, Nickell S, Baumeister W. Localization of protein complexes by pattern recognition. Methods Cell Biol. 2007;79:615\u201338.","journal-title":"Methods Cell Biol"},{"issue":"3","key":"4901_CR5","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1016\/j.jsb.2016.07.006","volume":"195","author":"F Wang","year":"2016","unstructured":"Wang F, Gong H, Liu G, Li M, Yan C, Xia T, Li X, Zeng J. DeepPicker: a deep learning approach for fully automated particle picking in cryo-EM. J Struct Biol. 2016;195(3):325\u201336.","journal-title":"J Struct Biol"},{"issue":"1","key":"4901_CR6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12859-016-1414-x","volume":"18","author":"Y Zhu","year":"2017","unstructured":"Zhu Y, Ouyang Q, Mao Y. A deep convolutional neural network approach to single-particle recognition in cryo-electron microscopy. BMC Bioinf. 2017;18(1):1\u201310.","journal-title":"BMC Bioinf"},{"issue":"11","key":"4901_CR7","doi-asserted-by":"publisher","first-page":"1153","DOI":"10.1038\/s41592-019-0575-8","volume":"16","author":"T Bepler","year":"2019","unstructured":"Bepler T, Morin A, Rapp M, Brasch J, Shapiro L, Noble AJ, Berger B. Positive-unlabeled convolutional neural networks for particle picking in cryo-electron micrographs. Nat Methods. 2019;16(11):1153\u201360.","journal-title":"Nat Methods"},{"issue":"1","key":"4901_CR8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s42003-019-0437-z","volume":"2","author":"T Wagner","year":"2019","unstructured":"Wagner T, Merino F, Stabrin M, Moriya T, Antoni C, Apelbaum A, Hagel P, Sitsel O, Raisch T, Prumbaum D, et al. SPHIRE-crYOLO is a fast and accurate fully automated particle picker for cryo-EM. Commun Biol. 2019;2(1):1\u201313.","journal-title":"Commun Biol"},{"key":"4901_CR9","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek \u00d6, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O. 3D U-Net: learning dense volumetric segmentation from sparse annotation. In: International conference on medical image computing and computer-assisted intervention, pp 424\u2013432 (2016). Springer","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"4901_CR10","doi-asserted-by":"crossref","unstructured":"Gubins I, van\u00a0der Schot G, Veltkamp RC, F\u00f6rster F, Du X, Zeng X, Zhu Z, Chang L, Xu M, Moebel E, et al. Classification in cryo-electron tomograms. SHREC\u201919 Track (2019)","DOI":"10.1016\/j.cag.2020.07.010"},{"issue":"14","key":"4901_CR11","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1093\/bioinformatics\/btx230","volume":"33","author":"M Xu","year":"2017","unstructured":"Xu M, Chai X, Muthakana H, Liang X, Yang G, Zeev-Ben-Mordehai T, Xing EP. Deep learning-based subdivision approach for large scale macromolecules structure recovery from electron cryo tomograms. Bioinformatics. 2017;33(14):13\u201322.","journal-title":"Bioinformatics"},{"key":"4901_CR12","unstructured":"Tran D, Bourdev LD, Fergus R, Torresani L, Paluri M. C3D: generic features for video analysis. CoRR, abs\/1412.0767 2(7), 8 (2014)"},{"key":"4901_CR13","unstructured":"Sabour S, Frosst N, Hinton GE. Dynamic routing between capsules. arXiv preprint arXiv:1710.09829 (2017)"},{"key":"4901_CR14","doi-asserted-by":"crossref","unstructured":"Mazzia V, Salvetti F, Chiaberge M. Efficient-CapsNet: Capsule network with self-attention routing. arXiv preprint arXiv:2101.12491 (2021)","DOI":"10.1038\/s41598-021-93977-0"},{"issue":"1","key":"4901_CR15","first-page":"1295","volume":"34","author":"MK Patrick","year":"2022","unstructured":"Patrick MK, Adekoya AF, Mighty AA, Edward BY. Capsule networks\u2014a survey. J King Saud Univ Comput Inf Sci. 2022;34(1):1295\u2013310.","journal-title":"J King Saud Univ Comput Inf Sci"},{"key":"4901_CR16","doi-asserted-by":"crossref","unstructured":"Zhao Y, Birdal T, Deng H, Tombari F. 3D point capsule networks. In: Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition, pp 1009\u20131018 (2019)","DOI":"10.1109\/CVPR.2019.00110"},{"key":"4901_CR17","doi-asserted-by":"crossref","unstructured":"Nguyen T, Hua B-S, Le N. 3D-UCaps: 3D capsules UNET for volumetric image segmentation. In: International conference on medical image computing and computer-assisted intervention, pp 548\u2013558 (2021). Springer","DOI":"10.1007\/978-3-030-87193-2_52"},{"key":"4901_CR18","doi-asserted-by":"crossref","unstructured":"Da T, Ding J, Yang L, Chirikjian G. A method for fully automated particle picking in cryo-electron microscopy based on a CNN. In: Proceedings of the 2018 ACM international conference on bioinformatics, computational biology, and health informatics, pp 633\u2013638 (2018)","DOI":"10.1145\/3233547.3233706"},{"issue":"1","key":"4901_CR19","doi-asserted-by":"publisher","first-page":"5364","DOI":"10.1038\/s41467-021-25413-w","volume":"12","author":"PS Erdmann","year":"2021","unstructured":"Erdmann PS, Hou Z, Klumpe S, Khavnekar S, Beck F, Wilfling F, Plitzko JM, Baumeister W. In situ cryo-electron tomography reveals gradient organization of ribosome biogenesis in intact nucleoli. Nat Commun. 2021;12(1):5364. https:\/\/doi.org\/10.1038\/s41467-021-25413-w.","journal-title":"Nat Commun"},{"issue":"S2","key":"4901_CR20","doi-asserted-by":"publisher","first-page":"2516","DOI":"10.1017\/S143192762002187X","volume":"26","author":"W Wan","year":"2020","unstructured":"Wan W, Khavnekar S, Wagner J, Erdmann P, Baumeister W. Stopgap: a software package for subtomogram averaging and refinement. Microsc Microanal. 2020;26(S2):2516\u20132516.","journal-title":"Microsc Microanal"}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-022-04901-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-022-04901-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-022-04901-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,30]],"date-time":"2022-08-30T21:11:30Z","timestamp":1661893890000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcbioinformatics.biomedcentral.com\/articles\/10.1186\/s12859-022-04901-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,30]]},"references-count":20,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,12]]}},"alternative-id":["4901"],"URL":"https:\/\/doi.org\/10.1186\/s12859-022-04901-w","relation":{},"ISSN":["1471-2105"],"issn-type":[{"type":"electronic","value":"1471-2105"}],"subject":[],"published":{"date-parts":[[2022,8,30]]},"assertion":[{"value":"24 June 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 August 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 August 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"360"}}