Fold'EM: Direct atomic structure inference from Cryo-EM particles
Abstract
Single-particle cryo-electron microscopy (cryo-EM) has become a widely adopted technique for biomolecular structure determination. The conventional cryo-EM computational pipeline first combines many particle images to reconstruct an electrostatic potential (ESP) map and then fits an atomic model to the recovered map. Density reconstruction has high sample complexity, requiring large numbers of particle images and making structure determination high-cost and low-throughput, particularly for heterogeneous samples. Downstream atomic model building, in turn, becomes increasingly difficult as the resolution of the reconstructed map deteriorates. Protein structure prediction models provide strong sequence-derived priors on atomic structure, and experiment-guided approaches can use these priors to recover structures consistent with experimental measurements. Yet, in cryo-EM, such priors are typically integrated only after density reconstruction during atomic model fitting. We introduce Fold'EM, an inference-time framework that combines priors from protein generative models directly with cryo-EM particle images to determine atomic models from a small number of single particle images, bypassing both intermediate density reconstruction and downstream model building against the reconstructed map. Across synthetic and experimental cryo-EM datasets, Fold'EM recovers accurate atomic structures both with known particle orientations and in an ab-initio setting where orientations are inferred jointly with structure. In heterogeneous datasets, Fold'EM further resolves distinct conformational states from mixed particle populations without separately reconstructing a density map and building an atomic model for each state. We believe these results open new avenues for structure determination in the low-sample regime and for characterizing low-population conformational states directly from cryo-EM particles.
Related
- Structure of a sparsely populated chimeric intermediate that facilitates fold-switching of a metamorphic protein
- The Rossmann2x2 Fold Attains its Native Structure Via a Defined Pathway of Sequential and Cooperative Folding Units
- Atomic structure and dynamics of the mechanosensitive channel MscL from Escherichia coli by cryo-EM and solid-state NMR
- DeCoRE: A computational method to resolve RNA structural heterogeneity from RNA structure probing data by direct RNA sequencing
- Context-dependent protein structure prediction analysis and complex stoichiometry inference with MultimerMapper.
BibTeX
@misc{maddipatla2026foldem,
title = {{Fold'EM: Direct atomic structure inference from Cryo-EM particles}},
author = {Advaith Maddipatla and Märt-Erik Mäeots and Marco Pegoraro and Nikolaus Dräger and Roberto Covino and Sanketh Vedula and Martin Pacesa and Alex M. Bronstein},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.01358},
eprint = {2610.01358},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.01358},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Fold'EM: Direct atomic structure inference from Cryo-EM particles
AU - Advaith Maddipatla
AU - Märt-Erik Mäeots
AU - Marco Pegoraro
AU - Nikolaus Dräger
AU - Roberto Covino
AU - Sanketh Vedula
AU - Martin Pacesa
AU - Alex M. Bronstein
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.01358
UR - https://arxiv.org/abs/2610.01358
AB - Single-particle cryo-electron microscopy (cryo-EM) has become a widely adopted technique for biomolecular structure determination. The conventional cryo-EM computational pipeline first combines many particle images to reconstruct an electrostatic potential (ESP) map and then fits an atomic model to the recovered map. Density reconstruction has high sample complexity, requiring large numbers of particle images and making structure determination high-cost and low-throughput, particularly for heterogeneous samples. Downstream atomic model building, in turn, becomes increasingly difficult as the resolution of the reconstructed map deteriorates. Protein structure prediction models provide strong sequence-derived priors on atomic structure, and experiment-guided approaches can use these priors to recover structures consistent with experimental measurements. Yet, in cryo-EM, such priors are typically integrated only after density reconstruction during atomic model fitting. We introduce Fold'EM, an inference-time framework that combines priors from protein generative models directly with cryo-EM particle images to determine atomic models from a small number of single particle images, bypassing both intermediate density reconstruction and downstream model building against the reconstructed map. Across synthetic and experimental cryo-EM datasets, Fold'EM recovers accurate atomic structures both with known particle orientations and in an ab-initio setting where orientations are inferred jointly with structure. In heterogeneous datasets, Fold'EM further resolves distinct conformational states from mixed particle populations without separately reconstructing a density map and building an atomic model for each state. We believe these results open new avenues for structure determination in the low-sample regime and for characterizing low-population conformational states directly from cryo-EM particles.
ER -