Overcoming the Preferred-Orientation Problem in Cryo-EM with Self-Supervised Deep Learning. Liu, Y., Fan, H., Hu, J. J., & Zhou, Z. H. Nature Methods, 22(1):113–123, Nature Publishing Group, January, 2025. doi abstract bibtex While advances in single-particle cryo-EM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the `preferred' orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep learning-based software to address map anisotropy and particle misalignment caused by the preferred-orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's ability to generate near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, $β$-galactosidases and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred-orientation problem.
@article{liuOvercomingPreferredorientationProblem2025,
title = {Overcoming the Preferred-Orientation Problem in Cryo-{{EM}} with Self-Supervised Deep Learning},
author = {Liu, Yun-Tao and Fan, Hongcheng and Hu, Jason J. and Zhou, Z. Hong},
year = {2025},
month = jan,
journal = {Nature Methods},
volume = {22},
number = {1},
pages = {113--123},
publisher = {Nature Publishing Group},
issn = {1548-7105},
doi = {10.1038/s41592-024-02505-1},
urldate = {2025-04-04},
abstract = {While advances in single-particle cryo-EM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the `preferred' orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep learning-based software to address map anisotropy and particle misalignment caused by the preferred-orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's ability to generate near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, {$\beta$}-galactosidases and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred-orientation problem.},
copyright = {2024 The Author(s), under exclusive licence to Springer Nature America, Inc.},
langid = {english},
keywords = {Cryoelectron microscopy,Cryoelectron tomography,Machine learning,Software},
file = {C:\Users\shervinnia\Zotero\storage\2WR59ZWV\Liu et al. - 2025 - Overcoming the preferred-orientation problem in cryo-EM with self-supervised deep learning.pdf}
}
Downloads: 0
{"_id":"aN9qG8HEnXsd5uBF4","bibbaseid":"liu-fan-hu-zhou-overcomingthepreferredorientationproblemincryoemwithselfsuperviseddeeplearning-2025","author_short":["Liu, Y.","Fan, H.","Hu, J. J.","Zhou, Z. H."],"bibdata":{"bibtype":"article","type":"article","title":"Overcoming the Preferred-Orientation Problem in Cryo-EM with Self-Supervised Deep Learning","author":[{"propositions":[],"lastnames":["Liu"],"firstnames":["Yun-Tao"],"suffixes":[]},{"propositions":[],"lastnames":["Fan"],"firstnames":["Hongcheng"],"suffixes":[]},{"propositions":[],"lastnames":["Hu"],"firstnames":["Jason","J."],"suffixes":[]},{"propositions":[],"lastnames":["Zhou"],"firstnames":["Z.","Hong"],"suffixes":[]}],"year":"2025","month":"January","journal":"Nature Methods","volume":"22","number":"1","pages":"113–123","publisher":"Nature Publishing Group","issn":"1548-7105","doi":"10.1038/s41592-024-02505-1","urldate":"2025-04-04","abstract":"While advances in single-particle cryo-EM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the `preferred' orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep learning-based software to address map anisotropy and particle misalignment caused by the preferred-orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's ability to generate near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, $β$-galactosidases and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred-orientation problem.","copyright":"2024 The Author(s), under exclusive licence to Springer Nature America, Inc.","langid":"english","keywords":"Cryoelectron microscopy,Cryoelectron tomography,Machine learning,Software","file":"C:\\Users\\shervinnia\\Zotero\\storage\\2WR59ZWVŁiu et al. - 2025 - Overcoming the preferred-orientation problem in cryo-EM with self-supervised deep learning.pdf","bibtex":"@article{liuOvercomingPreferredorientationProblem2025,\n title = {Overcoming the Preferred-Orientation Problem in Cryo-{{EM}} with Self-Supervised Deep Learning},\n author = {Liu, Yun-Tao and Fan, Hongcheng and Hu, Jason J. and Zhou, Z. Hong},\n year = {2025},\n month = jan,\n journal = {Nature Methods},\n volume = {22},\n number = {1},\n pages = {113--123},\n publisher = {Nature Publishing Group},\n issn = {1548-7105},\n doi = {10.1038/s41592-024-02505-1},\n urldate = {2025-04-04},\n abstract = {While advances in single-particle cryo-EM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the `preferred' orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep learning-based software to address map anisotropy and particle misalignment caused by the preferred-orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's ability to generate near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, {$\\beta$}-galactosidases and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred-orientation problem.},\n copyright = {2024 The Author(s), under exclusive licence to Springer Nature America, Inc.},\n langid = {english},\n keywords = {Cryoelectron microscopy,Cryoelectron tomography,Machine learning,Software},\n file = {C:\\Users\\shervinnia\\Zotero\\storage\\2WR59ZWV\\Liu et al. - 2025 - Overcoming the preferred-orientation problem in cryo-EM with self-supervised deep learning.pdf}\n}\n\n","author_short":["Liu, Y.","Fan, H.","Hu, J. J.","Zhou, Z. H."],"key":"liuOvercomingPreferredorientationProblem2025","id":"liuOvercomingPreferredorientationProblem2025","bibbaseid":"liu-fan-hu-zhou-overcomingthepreferredorientationproblemincryoemwithselfsuperviseddeeplearning-2025","role":"author","urls":{},"keyword":["Cryoelectron microscopy","Cryoelectron tomography","Machine learning","Software"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://bibbase.org/network/files/2MtabhipmJFGQ7n9n","dataSources":["mrv2RSxFMmgGBMu4y"],"keywords":["cryoelectron microscopy","cryoelectron tomography","machine learning","software"],"search_terms":["overcoming","preferred","orientation","problem","cryo","self","supervised","deep","learning","liu","fan","hu","zhou"],"title":"Overcoming the Preferred-Orientation Problem in Cryo-EM with Self-Supervised Deep Learning","year":2025}