This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| RELION | |
|---|---|
| Name | RELION |
| Developer | Shaun M. Scheres group, MRC Laboratory of Molecular Biology |
| Released | 2012 |
| Programming language | C++, Python (programming language) |
| Operating system | Unix-like |
| Genre | Cryo-electron microscopy software |
| License | GNU General Public License |
RELION RELION is a software package for single-particle cryo-electron microscopy data analysis that implements a Bayesian approach to three-dimensional reconstruction. It integrates statistical models and computational routines to perform particle alignment, classification, and refinement, and has been influential in work from institutions such as the MRC Laboratory of Molecular Biology, Max Planck Institute, Harvard University, Stanford University, and University of Cambridge. Widely used by laboratories involved with EMBL-EBI, National Institutes of Health, European Molecular Biology Laboratory, and cryo-EM centers at Caltech, RELION has contributed to structural determinations that intersect with projects at Broad Institute, Scripps Research, and national facilities like Diamond Light Source.
RELION is centered on a probabilistic framework that models cryo-EM images as noisy projections of unknown three-dimensional macromolecular structures; it uses expectation-maximization and empirical Bayesian inference to estimate parameters. The package was developed within a research environment linked to the MRC Laboratory of Molecular Biology and has been cited alongside seminal structural discoveries at institutions including University of Oxford, Yale University, University of California, Berkeley, Princeton University, and Columbia University. Researchers from facilities such as Argonne National Laboratory, Brookhaven National Laboratory, and Lawrence Berkeley National Laboratory have adopted RELION alongside complementary tools from projects at Cold Spring Harbor Laboratory and EMBL.
RELION originated in the early 2010s as a response to challenges seen in reconstruction projects at centers like MRC-LMB and groups led by investigators associated with Medical Research Council programs. Initial versions were developed by the group led by Shaun Scheres and collaborators at MRC Laboratory of Molecular Biology, with algorithmic advances communicated through conferences such as the Gordon Research Conference and symposiums at ICMS and EMBO. Adoption accelerated following high-profile structural publications from teams at Max Planck Institute for Biophysical Chemistry, Cold Spring Harbor Laboratory, Johns Hopkins University, and University of Toronto, and subsequent expansions incorporated GPU acceleration influenced by developments at NVIDIA research collaborations and computing centers like Argonne Leadership Computing Facility.
RELION's architecture combines C++ core routines and Python-based wrappers to orchestrate dataflow and job management across clusters and workstations. Its core implements maximum a posteriori estimation with expectation-maximization, probabilistic classification, and regularization strategies; these methods relate conceptually to statistical work from groups at Stanford University, MIT, University of Cambridge, University of Oxford, and algorithmic developments that have been shared at venues such as NeurIPS and ISMB. Computational performance has been bolstered through integration with GPU toolchains originating in collaboration with teams at NVIDIA, and deployment workflows often leverage scheduling systems used at Lawrence Berkeley National Laboratory and Oak Ridge National Laboratory.
Typical RELION workflows encompass motion correction, contrast transfer function estimation, particle picking, 2D classification, 3D classification, and auto-refinement; users often combine RELION steps with preprocessing tools from projects at UCSF, University of California, San Francisco, EMBL-EBI, and University of Heidelberg. The package supports micrograph import, metadata bookkeeping, and job pipelining compatible with cluster managers used at Princeton University and visualization with viewers adopted by groups at Weizmann Institute of Science and University of Tokyo. Key features include hierarchical classification schemes that echo theoretical approaches from ETH Zurich and regularized refinement strategies that mirror statistical practices from Columbia University and University of Chicago.
RELION's accuracy has been benchmarked against datasets produced in laboratories at MRC Laboratory of Molecular Biology, Max Planck Institute, University of Cambridge, and national electron microscopy facilities such as eBIC at Diamond Light Source. Validation protocols frequently reference standards developed in consortia involving EMBL, EMBO, and National Institutes of Health. Performance scaling studies report speedups using GPU-accelerated routines compatible with hardware promoted by NVIDIA and cluster configurations similar to those at Argonne National Laboratory and Oak Ridge National Laboratory, while community-driven validation efforts connect to initiatives at EMDataResource and projects hosted by EMBL-EBI.
RELION has been applied to determine structures of ribosomes, ion channels, membrane complexes, and viral capsids studied by teams at Harvard Medical School, Princeton University, Scripps Research, Rockefeller University, University of California, San Diego, and Yale University. Its role in resolving conformational heterogeneity has supported mechanistic studies carried out in laboratories at Max Planck Institute for Biochemistry, ETH Zurich, University of Toronto, and Columbia University. Collaborative projects combining RELION output with integrative modeling approaches from European Molecular Biology Laboratory, Broad Institute, and University College London have informed functional hypotheses tested in biochemical labs at Johns Hopkins University and University of British Columbia.
RELION is distributed under the GNU General Public License with source contributions coordinated through repositories and mirrors used by research groups at MRC Laboratory of Molecular Biology, EMBL-EBI, Max Planck Society, and community developers affiliated with University of Cambridge and University of Oxford. Precompiled binaries and container images are available for Unix-like systems commonly used at institutions such as Caltech, Stanford University, and Harvard University, and community support is provided via forums and workshops hosted by organizations including EMBO and BioEM.
Category:Cryo-electron microscopy software