← Back to all news
arXiv Machine Learning By Nithin Chalapathi, Theo Sternlieb, Ethan Han, Sonya M. Hanson, Pilar Cossio, Aditi S. Krishnapriyan, Ritwik Gupta

Learning the Shared Latent Structure of Cryo-EM Inverse Problems

Read the original on arXiv Machine Learning →

The paper introduces EMber, a family of vision transformers trained on over 50 million cryo‑EM particle images and simulated projections, to learn a shared latent representation for multiple inverse problems in cryo‑EM. By employing self‑distillation with global and masked patch objectives, EMber captures spatially localized structural differences and improves performance on conformational and compositional discrimination by roughly 15 percentage points compared to existing methods. The authors demonstrate that these representations enable accurate estimation of conformational populations and yield high‑quality reconstructions in fewer training epochs.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Sep 30

Cryo-EM as a Stochastic Inverse Problem

arXiv:2509.05541v2 Announce Type: replace-cross Abstract: Cryo-electron microscopy (Cryo-EM) enables high-resolution imaging of biomolecules, but structural heterogeneity remains a major challenge in...

By Diego Sanchez Espinosa, Erik H Thiede, Yunan Yang
More like this →
arXiv Machine Learning
Sep 7

The microscope is the mask: privileged views and labels from a cryo-ET forward model

The paper introduces CARNIVAL, a model for protein annotation in cryo-electron tomography (cryo-ET) volumes that leverages simulated data and a forward model to generate domain‑specific augmented paired views for self‑supervised training. By incorporating simulation‑derived protein positions and identities into the architecture and loss function, the model localises semantic information at protein locations. CARNIVAL is evaluated on real tomograms without finetuning and outperforms a state‑of‑the‑art contrastive model that lacks forward‑model paired views or privileged information.

By Bogdan Toader, Kiarash Jamali, Tanmay A. M. Bharat, Sjors H. W. Scheres
fine-tuningbenchmarks
More like this →
arXiv AI
Oct 2

Fold'EM: Direct atomic structure inference from Cryo-EM particles

Fold'EM is an inference-time framework that directly combines protein generative model priors with cryo‑EM particle images to recover atomic structures, bypassing the traditional density reconstruction step. It can determine structures from a small number of particles, jointly infer orientations in an ab‑initio setting, and resolve distinct conformational states in heterogeneous samples without separate map reconstruction. The method demonstrates accurate atomic models on both synthetic and experimental datasets, showing promise for low‑sample and low‑population conformational analysis.

By Advaith Maddipatla, M\"art-Erik M\"aeots, Marco Pegoraro, Nikolaus Dr\"ager, Roberto Covino, Sanketh Vedula, Martin Pacesa, Alex M. Bronstein
efficiency
More like this →
arXiv Machine Learning
Jun 10

POPSICLE: Benchmark Datasets for Segmentation and Localization in CryoET

arXiv:2606. 10255v1 Announce Type: cross Abstract: Cryo-electron tomography (cryoET) has emerged as a powerful tool in structural and cellular biology by enabling direct visualization of macromolecular structures within intact cells, thereby linking molecular architecture to cellular organization in a native context.

By Jonathan Schwartz, Utz Heinrich Ermel, C. Braxton Owens, Zhuowen Zhao, Ariana Peck, Gus L. W. Hart, Grant J. Jensen, Bridget Carragher, Dari Kimanius
computer-visionbenchmarks
More like this →
arXiv Machine Learning
Oct 9

Learning from Hetero Density for Cryo-EM Protein Reconstruction

The paper presents CryoCue, a framework that leverages hetero component information to enhance protein reconstruction from cryo-electron microscopy maps. An anchor‑supervised detector learns representations for five hetero classes, and multiscale hetero features guide backbone localization while predicted hetero candidates inform structure refinement. Experiments demonstrate that CryoCue improves backbone localization near hetero components and yields more accurate protein structures.

By Xu Han, Chaozhuo Li, Xiaowei Yuan, Yuancheng Sun, Kang Liu, Qiwei Ye
More like this →