Browse
Submit Data
Databases
API
Help

Dataset Information

45 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Machine learning-based real-time object locator/evaluator for cryo-EM data collection.

ABSTRACT: In cryo-electron microscopy (cryo-EM) data collection, locating a target object is error-prone. Here, we present a machine learning-based approach with a real-time object locator named yoneoLocr using YOLO, a well-known object detection system. Implementation shows its effectiveness in rapidly and precisely locating carbon holes in single particle cryo-EM and in locating crystals and evaluating electron diffraction (ED) patterns in automated cryo-electron crystallography (cryo-EX) data collection. The proposed approach will advance high-throughput and accurate data collection of images and diffraction patterns with minimal human operation.

SUBMITTER: Yonekura K

PROVIDER: S-EPMC8423793 | biostudies-literature | 2021 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

Machine learning-based real-time object locator/evaluator for cryo-EM data collection.

Yonekura Koji K Maki-Yonekura Saori S Naitow Hisashi H Hamaguchi Tasuku T Takaba Kiyofumi K

Communications biology 20210907 1

In cryo-electron microscopy (cryo-EM) data collection, locating a target object is error-prone. Here, we present a machine learning-based approach with a real-time object locator named yoneoLocr using YOLO, a well-known object detection system. Implementation shows its effectiveness in rapidly and precisely locating carbon holes in single particle cryo-EM and in locating crystals and evaluating electron diffraction (ED) patterns in automated cryo-electron crystallography (cryo-EX) data collectio ...[more]

PMID: 34493805

Similar Datasets

SIMPLE 3.0. Stream single-particle cryo-EM analysis in real time.

Project description:We here introduce the third major release of the SIMPLE (Single-particle IMage Processing Linux Engine) open-source software package for analysis of cryogenic transmission electron microscopy (cryo-EM) movies of single-particles (Single-Particle Analysis, SPA). Development of SIMPLE 3.0 has been focused on real-time data processing using minimal CPU computing resources to allow easy and cost-efficient scaling of processing as data rates escalate. Our stream SPA tool implements the steps of anisotropic motion correction and CTF estimation, rapid template-based particle identification and 2D clustering with automatic class rejection. SIMPLE 3.0 additionally features an easy-to-use web-based graphical user interface (GUI) that can be run on any device (workstation, laptop, tablet or phone) and supports a remote multi-user environment over the network. The new project-based execution model automatically records the executed workflow and represents it as a flow diagram in the GUI. This facilitates meta-data handling and greatly simplifies usage. Using SIMPLE 3.0, it is possible to automatically obtain a clean SP data set amenable to high-resolution 3D reconstruction directly upon completion of the data acquisition, without the need for extensive image processing post collection. Only minimal standard CPU computing resources are required to keep up with a rate of ∼300 Gatan K3 direct electron detector movies per hour. SIMPLE 3.0 is available for download from simplecryoem.com.

| S-EPMC7695977 | biostudies-literature

Cryo-EM structure of a 3D DNA-origami object.

Project description:A key goal for nanotechnology is to design synthetic objects that may ultimately achieve functionalities known today only from natural macromolecular complexes. Molecular self-assembly with DNA has shown potential for creating user-defined 3D scaffolds, but the level of attainable positional accuracy has been unclear. Here we report the cryo-EM structure and a full pseudoatomic model of a discrete DNA object that is almost twice the size of a prokaryotic ribosome. The structure provides a variety of stable, previously undescribed DNA topologies for future use in nanotechnology and experimental evidence that discrete 3D DNA scaffolds allow the positioning of user-defined structural motifs with an accuracy that is similar to that observed in natural macromolecules. Thereby, our results indicate an attractive route to fabricate nanoscale devices that achieve complex functionalities by DNA-templated design steered by structural feedback.

| S-EPMC3523823 | biostudies-literature

Overcoming resolution attenuation during tilted cryo-EM data collection.

Project description:Structural biology efforts using cryogenic electron microscopy are frequently stifled by specimens adopting "preferred orientations" on grids, leading to anisotropic map resolution and impeding structure determination. Tilting the specimen stage during data collection is a generalizable solution but has historically led to substantial resolution attenuation. Here, we develop updated data collection and image processing workflows and demonstrate, using multiple specimens, that resolution attenuation is negligible or significantly reduced across tilt angles. Reconstructions with and without the stage tilted as high as 60° are virtually indistinguishable. These strategies allowed the reconstruction to 3 Å resolution of a bacterial RNA polymerase with preferred orientation, containing an unnatural nucleotide for studying novel base pair recognition. Furthermore, we present a quantitative framework that allows cryo-EM practitioners to define an optimal tilt angle during data acquisition. These results reinforce the utility of employing stage tilt for data collection and provide quantitative metrics to obtain isotropic maps.

| S-EPMC10776679 | biostudies-literature

Light-coupled cryo-plunger for time-resolved cryo-EM.

Project description:Proteins are dynamic molecules that can undergo rapid conformational rearrangements in response to stimuli. These structural changes are often critical to protein function, and thus elucidating time-dependent conformational landscapes has been a long-standing goal of structural biology. To harness the power of single particle cryo-EM methods to enable 'time-resolved' structure determination, we have developed a light-coupled cryo-plunger that pairs flash-photolysis of caged ligands with rapid sample vitrification. The 'flash-plunger' consists of a high-power ultraviolet LED coupled with focusing optics and a motorized linear actuator, enabling the user to immobilize protein targets in vitreous ice within a programmable time window - as short as tens of milliseconds - after stimulus delivery. The flash-plunger is a simple, inexpensive and flexible tool to explore short-lived conformational states previously unobtainable by conventional sample preparation methods.

| S-EPMC7959588 | biostudies-literature

Time-resolved cryo-EM using Spotiton.

Project description:We present an approach for preparing cryo-electron microscopy (cryo-EM) grids to study short-lived molecular states. Using piezoelectric dispensing, two independent streams of ~50-pl droplets of sample are deposited within 10 ms of each other onto the surface of a nanowire EM grid, and the mixing reaction stops when the grid is vitrified in liquid ethane ~100 ms later. We demonstrate this approach for four biological systems where short-lived states are of high interest.

| S-EPMC7799389 | biostudies-literature

A large expert-curated cryo-EM image dataset for machine learning protein particle picking.

Project description:Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the structures of biological macromolecular complexes. Picking single-protein particles from cryo-EM micrographs is a crucial step in reconstructing protein structures. However, the widely used template-based particle picking process is labor-intensive and time-consuming. Though machine learning and artificial intelligence (AI) based particle picking can potentially automate the process, its development is hindered by lack of large, high-quality labelled training data. To address this bottleneck, we present CryoPPP, a large, diverse, expert-curated cryo-EM image dataset for protein particle picking and analysis. It consists of labelled cryo-EM micrographs (images) of 34 representative protein datasets selected from the Electron Microscopy Public Image Archive (EMPIAR). The dataset is 2.6 terabytes and includes 9,893 high-resolution micrographs with labelled protein particle coordinates. The labelling process was rigorously validated through 2D particle class validation and 3D density map validation with the gold standard. The dataset is expected to greatly facilitate the development of both AI and classical methods for automated cryo-EM protein particle picking.

| S-EPMC10287764 | biostudies-literature

Coma-corrected rapid single-particle cryo-EM data collection on the CRYO ARM 300.

Project description:Single-particle cryogenic electron microscopy has recently become a major method for determining the structures of proteins and protein complexes. This has markedly increased the demand for throughput of high-resolution electron microscopes, which are required to produce high-resolution images at high rates. An increase in data-collection throughput can be achieved by using large beam-image shifts combined with off-axis coma correction, enabling the acquisition of multiple images from a large area of the EM grid without moving the microscope stage. Here, the optical properties of the JEOL CRYO ARM 300 electron microscope equipped with a K3 camera were characterized under off-axis illumination conditions. It is shown that efficient coma correction can be achieved for beam-image shifts with an amplitude of at least 10 µm, enabling a routine throughput for data collection of between 6000 and 9000 images per day. Use of the benchmark for the rapid data-collection procedure (with beam-image shifts of up to 7 µm) on apoferritin resulted in a reconstruction at a resolution of 1.7 Å. This demonstrates that the rapid automated acquisition of high-resolution micrographs is possible using a CRYO ARM 300.

| S-EPMC8098478 | biostudies-literature

CryoPPP: A Large Expert-Labelled Cryo-EM Image Dataset for Machine Learning Protein Particle Picking.

Project description:Cryo-electron microscopy (cryo-EM) is currently the most powerful technique for determining the structures of large protein complexes and assemblies. Picking single-protein particles from cryo-EM micrographs (images) is a key step in reconstructing protein structures. However, the widely used template-based particle picking process is labor-intensive and time-consuming. Though the emerging machine learning-based particle picking can potentially automate the process, its development is severely hindered by lack of large, high-quality, manually labelled training data. Here, we present CryoPPP, a large, diverse, expert-curated cryo-EM image dataset for single protein particle picking and analysis to address this bottleneck. It consists of manually labelled cryo-EM micrographs of 32 non-redundant, representative protein datasets selected from the Electron Microscopy Public Image Archive (EMPIAR). It includes 9,089 diverse, high-resolution micrographs (âˆ¼300 cryo-EM images per EMPIAR dataset) in which the coordinates of protein particles were labelled by human experts. The protein particle labelling process was rigorously validated by both 2D particle class validation and 3D density map validation with the gold standard. The dataset is expected to greatly facilitate the development of machine learning and artificial intelligence methods for automated cryo-EM protein particle picking. The dataset and data processing scripts are available at https://github.com/BioinfoMachineLearning/cryoppp.

| S-EPMC9980126 | biostudies-literature

Single-Particle Cryo-EM Data Collection with Stage Tilt using Leginon.

Project description:Single-particle analysis (SPA) by cryo-electron microscopy (cryo-EM) is now a mainstream technique for high-resolution structural biology. Structure determination by SPA relies upon obtaining multiple distinct views of a macromolecular object vitrified within a thin layer of ice. Ideally, a collection of uniformly distributed random projection orientations would amount to all possible views of the object, giving rise to reconstructions characterized by isotropic directional resolution. However, in reality, many samples suffer from preferentially oriented particles adhering to the air-water interface. This leads to non-uniform angular orientation distributions in the dataset and inhomogeneous Fourier-space sampling in the reconstruction, translating into maps characterized by anisotropic resolution. Tilting the specimen stage provides a generalizable solution to overcoming resolution anisotropy by virtue of improving the uniformity of orientation distributions, and thus the isotropy of Fourier space sampling. The present protocol describes a tilted-stage automated data collection strategy using Leginon, a software for automated image acquisition. The procedure is simple to implement, does not require any additional equipment or software, and is compatible with most standard transmission electron microscopes (TEMs) used for imaging biological macromolecules.

| S-EPMC9733953 | biostudies-literature

MealTime-MS: A Machine Learning-Guided Real-Time Mass Spectrometry Analysis for Proteomics

Project description:Mass spectrometry-based proteomics technologies are the prime methods for the high-throughput identification of proteins expressed in complex biological samples. Nevertheless, mass spectrometry’s technical limitations still hinder its ability to identify low abundance proteins in complex samples. Characterizing such proteins is essential to provide a comprehensive understanding of the biological processes taking place in a sample. Still today, a large part of the mass spectrometry-based proteomics performed use a data-dependent approach that favors the acquisition of mass spectra and detection of proteins of higher abundance. Combined to the fact that the computational identification of proteins from mass spectrometry data is typically performed after mass spectrometry data acquisition, large numbers of mass spectra are redundantly collected from the same abundant proteins and little to no mass spectra are acquired for proteins of lower abundance. To address this problem, we propose a novel supervised learning algorithm that identifies proteins in real-time as mass spectrometry data is acquired and prevents the further data acquisition related to confidently identified proteins to improve the identification sensitivity of low abundance proteins. We show in real-time simulations of a previously performed mass spectrometry analysis of a HEK293 cell lysates that our approach can identify 92.1% of the proteins using 66.2% of the MS2 spectra acquired in the experiment. We also demonstrate that our approach is fast enough for real-time mass spectrometry analysis, is flexible and that it outperforms previously proposed methods. Our method efficient usage of mass spectrometry resources will provide a more comprehensive characterization of proteomes in complex samples.

2020-10-27 | PXD017673 | Pride

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data