Advanced Electron Microscopy Techniques and Applications

This paper introduces a deep learning method to improve single particle cryoEM data quality by automatically filtering noisy images using denoising and domain adaptation neural networks.

CryoSiam is a self-supervised framework for cryo-ET data analysis that learns hierarchical representations and transfers to experimental data for denoising, segmentation, and macromolecular detection.

The paper studies the structural arrangement of the influenza C virus surface, finding hexagonal HEF lattice defects that align with theoretical predictions for curved lattices.

The paper uses electron cryo-tomography to determine the architecture of the Tad pilus in Caulobacter vibrioides and proposes an assembly model.

The paper introduces a parameter-free method for analyzing colocalization in cryo-electron tomograms to study spatial organization of protein complexes.

This paper introduces two optical geometries for correlative 4D-STEM and EELS to enable simultaneous structural and chemical analysis of biological specimens at sub-nanometre resolution.

The paper develops a method to express and visualize eukaryotic proteins using cell-free systems and cryo-electron tomography, demonstrating its application to study ASC protein filaments and their interaction with ribosomes.

Icecream improves cryo-ET reconstruction by addressing denoising, missing-wedge recovery, and computational efficiency using equivariant deep learning.

This paper automates repeat-target cryo-EM structure determination in CryoSPARC, achieving high-resolution results for 21 GPCR datasets with minimal human intervention.

The paper compares synapse detection in Drosophila EM reconstructions using SS-TEM and FIB-SEM, finding FIB-SEM detects more synapses.

The paper presents a 3.3 Å cryo-EM map of Vibrio cholerae competence pilus and a full-length pseudoatomic model of the CPM, revealing unique structural features and mechanical properties.

This paper optimizes adhesion methods for cryo-electron tomography of vitreous sections, increasing throughput to match thin film methods and enabling high-resolution imaging of molecular complexes.

This paper introduces a method to encapsulate individual C. elegans for TEM using EVOH hollow fibers, preserving specimen integrity and improving sectioning reproducibility.

ANTIDOTE improves cryoEM particle sorting by using metadata-driven neural networks, enhancing reconstruction quality and reducing manual curation efforts.

This paper develops a few-shot learning framework for cryo-EM micrograph classification using triplet loss embeddings, achieving competitive performance with minimal labeled examples.

This paper develops a deep learning method to correct non-uniform milling artifacts in FIB-SEM images using unsupervised cross-plane image-to-image translation.

The paper introduces an alternating angle milling technique to reduce streaking artifacts in FIB-SEM imaging, improving structural data quality by minimizing non-uniform material removal.

The paper proposes using electron backscatter diffraction in a cryo-SEM to assess ice crystallinity in cryo-TEM samples before preparation, enabling better targeting of sample preparation steps.

The paper identifies and characterizes a new conserved protein complex, CAGE, found in both prokaryotes and eukaryotes, revealing its structural features and evolutionary distribution.

CryoFSL improves protein particle picking in cryo-EM micrographs with minimal annotations, achieving high accuracy and efficiency in low-SNR conditions.

LieOTMap-FFT aligns cryo-EM maps with atomic models using Lie algebra and FFT-accelerated optimal transport, achieving high accuracy in rigid body fitting.

LieOTMap fits high-resolution atomic models into cryo-EM maps using a differentiable framework based on Lie algebra and optimal transport, achieving accurate alignment of E. coli GroEL.

This paper introduces a combined MSI and cryo-ET pipeline for reliable single-cell classification in cryo-ET samples, enabling post-acquisition cell identification and classification.

LieMap fits high-resolution atomic models into cryo-EM maps using gradient-based optimization, achieving a low RMSD for E. coli GroEL structure fitting.

RobPicker improves macromolecule identification in cryo-ET by addressing label noise and class imbalance through a meta-learning framework, outperforming existing methods in detecting small and rare species.

This paper introduces a method to overcome preferred orientation in cryo-EM by using ultrasonic excitation during vitrification, improving data acquisition efficiency.

This paper introduces c-SerialED, a method for high-quality protein structure determination using continuous serial electron diffraction, achieving atomic resolution with reduced radiation damage.

CryoDataBot automates the creation of high-quality cryoEM datasets for AI-driven structural biology, improving training efficiency and enabling flexible dataset customization.

This paper presents a workflow for cryo-EM structure determination of small proteins using heterogeneous ab initio reconstruction, successfully solving structures of three small proteins with resolutions below 5 Å.

The paper improves the resolution of the Cryo-EM structure of E. coli type I pilus, enabling detailed analysis of water molecule interactions and subunit stabilization.