{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T11:05:35Z","timestamp":1789297535296,"version":"build-2803163510"},"posted":{"date-parts":[[2026,9,10]]},"group-title":"Biophysics","reference-count":26,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"name":"SNF","award":["200020_192190"],"award-info":[{"award-number":["200020_192190"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Rare but biologically important molecular species are easily lost when reconstruction-based cryo-EM classification is applied to heterogeneous samples dominated by more abundant particles. We developed CryoConvNeXt, a deep-learning classifier that combines cyclic equivariance with adaptive frequency filtering. It is trained on simulated projections and adapted to experimental data by self-training, an unsupervised domain adaptation technique where the model acts as its own annotator. We tested Cryo-ConvNeXt on cryo-EM data collected for this study from controlled binary and ternary mixtures of Catalase, Apoferritin, and HSP60. Manual particle curation provided reference labels for benchmarks with class ratios from 1:1 to 1:16. CryoConvNeXt retained minority-species recall across this range. In contrast, cryoSPARC\u2019s recall collapsed in four of six pairwise conditions, as early as 1:4 when Catalase was the minority species. By recovering low-abundance species that conventional classification fails to recover reliably, CryoConvNeXt takes an important step towards solving a central problem in quantitative visual proteomics: measuring the molecular composition of complex biological samples.<\/jats:p>","DOI":"10.64898\/2026.09.08.750038","type":"posted-content","created":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T19:15:25Z","timestamp":1789067725000},"source":"Crossref","is-referenced-by-count":0,"title":["CryoConvNeXt enables robust identification of low-abundance molecular species in experimental cryo-EM data"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-5138-2501","authenticated-orcid":false,"given":"Larissa","family":"Glass","sequence":"first","affiliation":[{"name":"Biozentrum, University of Basel, 4056 Basel, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8216-1868","authenticated-orcid":false,"given":"Jan Pieter","family":"Abrahams","sequence":"additional","affiliation":[{"name":"Biozentrum, University of Basel, 4056 Basel, Switzerland"},{"name":"Laboratory of Nanoscale Biology, Paul Scherrer Institut 5232 Villigen PSI, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9148-0200","authenticated-orcid":false,"given":"Thomas","family":"Braun","sequence":"additional","affiliation":[{"name":"Biozentrum, University of Basel, 4056 Basel, Switzerland"},{"name":"Laboratory of Biological Electron Microscopy, Institute of Physics, School of Basic Science, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), 1015 Lausanne, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026091303300497000_2026.09.08.750038v1.1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2016.11.002"},{"key":"2026091303300497000_2026.09.08.750038v1.2","doi-asserted-by":"publisher","DOI":"10.1002\/pmic.201700176"},{"issue":"11","key":"2026091303300497000_2026.09.08.750038v1.3","doi-asserted-by":"crossref","first-page":"1153","DOI":"10.1038\/s41592-019-0575-8","volume":"16","year":"2019","journal-title":"Nature Methods"},{"issue":"8","key":"2026091303300497000_2026.09.08.750038v1.4","doi-asserted-by":"crossref","first-page":"930","DOI":"10.1038\/s41592-021-01220-5","volume":"18","year":"2021","journal-title":"Nature Methods"},{"key":"2026091303300497000_2026.09.08.750038v1.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2012.02.047"},{"key":"2026091303300497000_2026.09.08.750038v1.6","unstructured":"Cohen, T. & Welling, M. (2016). In Proceedings of The 33rd International Conference on Machine Learning, pp. 2990\u20132999. PMLR."},{"key":"2026091303300497000_2026.09.08.750038v1.7","doi-asserted-by":"publisher","DOI":"10.1021\/ac4027803"},{"key":"2026091303300497000_2026.09.08.750038v1.8","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.35383"},{"key":"2026091303300497000_2026.09.08.750038v1.9","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2013.06.012"},{"key":"2026091303300497000_2026.09.08.750038v1.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.sbi.2024.102815"},{"key":"2026091303300497000_2026.09.08.750038v1.11","doi-asserted-by":"crossref","unstructured":"Kopylov, M. , Kelley, K. , Yen, LY. , Rice, WJ. , Eng, ET. , Carragher, B. & Potter, CS . (2019). Horse spleen apoferritin light chain structure at 2.1 angstrom resolution.","DOI":"10.2210\/pdb6pxm\/pdb"},{"key":"2026091303300497000_2026.09.08.750038v1.12","unstructured":"Levy, A. , Grzadkowski, M. , Poitevin, F. , Vallese, F. , Clarke, O. B. , Wetzstein, G. & Zhong, E. D. (2024a). Revealing biomolecular structure and motion with neural ab initio cryo-em reconstruction."},{"key":"2026091303300497000_2026.09.08.750038v1.13","doi-asserted-by":"crossref","unstructured":"Levy, A. , Raghu, R. , Shustin, D. , Peng, A. R.-Y. , Li, H. , Clarke, O. B. , Wetzstein, G. & Zhong, E. D. (2024b). Mixture of neural fields for heterogeneous reconstruction in cryo-em.","DOI":"10.52202\/079017-1816"},{"key":"2026091303300497000_2026.09.08.750038v1.14","unstructured":"Liu, Z. , Mao, H. , Wu, C.-Y. , Feichtenhofer, C. , Darrell, T. & Xie, S. (2022). In Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, pp. 11976\u201311986."},{"key":"2026091303300497000_2026.09.08.750038v1.15","unstructured":"Loshchilov, I. & Hutter, F. (2019). Decoupled weight decay regularization."},{"issue":"12","key":"2026091303300497000_2026.09.08.750038v1.16","doi-asserted-by":"crossref","first-page":"2464","DOI":"10.1016\/j.cmet.2021.11.001","volume":"33","year":"2021","journal-title":"Cell Metabolism"},{"key":"2026091303300497000_2026.09.08.750038v1.17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2021.107702"},{"issue":"3","key":"2026091303300497000_2026.09.08.750038v1.18","doi-asserted-by":"crossref","first-page":"290","DOI":"10.1038\/nmeth.4169","volume":"14","year":"2017","journal-title":"Nature Methods"},{"key":"2026091303300497000_2026.09.08.750038v1.19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2015.08.008"},{"key":"2026091303300497000_2026.09.08.750038v1.20","doi-asserted-by":"publisher","DOI":"10.1111\/j.1365-2818.2011.03497.x"},{"key":"2026091303300497000_2026.09.08.750038v1.21","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2012.09.006"},{"key":"2026091303300497000_2026.09.08.750038v1.22","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmb.2005.02.031"},{"key":"2026091303300497000_2026.09.08.750038v1.23","unstructured":"Slowik, D. , Li, J. , Wilson, H. , Shtyrov, A. , Chen, S. , McMullan, G. , Russo, CJ. , Murshudov, G. & Henderson, R. (2025). Comparison of human and bacterial monofunctional catalase structures obtained by electron cryomicroscopy."},{"key":"2026091303300497000_2026.09.08.750038v1.24","unstructured":"Syntychaki, A. , Rima, L. , Schmidli, C. , Stohler, T. , Bieri, A. , S\u00fctterlin, R. , Stahlberg, H. , Casta\u00f1o-D\u00edez, D. & Braun, T. (2019). J. Proteome Res. p. 11."},{"key":"2026091303300497000_2026.09.08.750038v1.25","doi-asserted-by":"crossref","unstructured":"Szegedy, C. , Vanhoucke, V. , Ioffe, S. , Shlens, J. & Wojna, Z. (2016). In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2818\u20132826.","DOI":"10.1109\/CVPR.2016.308"},{"key":"2026091303300497000_2026.09.08.750038v1.26","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.453"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.08.750038","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T10:30:14Z","timestamp":1789295414000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.09.08.750038"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,10]]},"references-count":26,"URL":"https:\/\/doi.org\/10.64898\/2026.09.08.750038","relation":{},"subject":[],"published":{"date-parts":[[2026,9,10]]},"subtype":"preprint"}}