SPREAD
SPREAD performs automated end-to-end single-particle cryogenic electron microscopy (cryo-EM) data processing to generate three-dimensional reconstructions of biomacromolecules.
Key Features:
- Integration Across Processing Stages: Encompasses 2D image classification, 3D initial model generation, model selection, and 3D refinement within a single pipeline.
- Network-Based Image Classification (NCEM): Uses the NCEM network-based clustering algorithm for efficient and robust 2D image classification.
- Automatic Model Ranking and Selection (SRes): Implements the reference-free resolution measurement SRes to rank and select models automatically.
- Probabilistic Orientation Searching ("peak finding"): Applies a probabilistic orientation-searching method termed "peak finding" that leverages a network-based image similarity metric for projection orientation assignment.
- Mixture Image Network for Refinement: Constructs a mixture image network from peak-finding results to handle particle images and projection images during 3D refinement and recomputes node-pair similarities via a superposed random walk.
- Fully Automated Workflow: Executes the processing pipeline in a fully automated manner to minimize manual intervention during 2D classification, model selection, and 3D refinement.
Scientific Applications:
- 3D structure determination of biomacromolecules: Generates three-dimensional reconstructions of biomacromolecules from single-particle cryo-EM data for structural biology studies.
- Cryo-EM classification and refinement: Provides high-precision 2D classification, reference-free model ranking, and 3D refinement to support structural analysis and model selection.
Methodology:
Performs 2D image classification using NCEM; generates 3D initial models; ranks and selects models with SRes; assigns orientations via probabilistic "peak finding" using a network-based image similarity metric; constructs a mixture image network from peak-finding results and recomputes node-pair similarities through a superposed random walk for 3D refinement handling particle and projection images.
Topics
Details
- Added:
- 1/18/2021
- Last Updated:
- 2/21/2021
Operations
Publications
Xie R, Chen Y, Cai J, Yang Y, Shen H. SPREAD: A Fully Automated Toolkit for Single-Particle Cryogenic Electron Microscopy Data 3D Reconstruction with Image-Network-Aided Orientation Assignment. Journal of Chemical Information and Modeling. 2020;60(5):2614-2625. doi:10.1021/acs.jcim.9b01099. PMID:31990536.