The Electron Microscopy Data Bank

Lead Research Organisation: University of Dundee
Department Name: College of Life Sciences

Abstract

Cryo-Electron Microscopy, a microscopy technique in which electrons play the role of light in conventional microscopes, is a central technology in attempting to understand the inner workings of the living cell. It enables researchers to obtain images of cells and molecules, instantaenously frozen and embedded in an extremely thin layer of non-crystalline ice, at an extremely high resolution, at which the inner workings of the molecular machines that drive them can be discerned. The Electron Microscopy Data Bank (EMDB) is the central data bank for storing and distributing the three-dimensional images taken using advanced electron microscopy techniques with biological samples. Initiated in 2002 by Dr Kim Henrick, the head of the Macromolecular Structure Database group of the European Bioinformatics Institute (EBI) in Cambridge, it already contains over 500 structures of macromolecular objects, and is expected to rapidly grow over the coming years, as improvements in microscope and specimen-preparation technology take effect. It provides structural biologists world-wide with a resource for the deposition, distribution, and analysis of such images to facilitate their research and further the understanding of the basic mechanisms of life, which as an ultimate goal will contribute to an improved understanding of the mechanisms of health and disease in plants, animals, and humans. The data bank is operated as an interactive web site at which researchers submit their data for deposition. To maintain the high quality of the data, every submission is curated by an experienced curator, who supports the researches in matters of quality and technical correctness, and then made publicly available through a simple, user-friendly web site. Particular care is taken to collect accurate metadata, i.e. information about the experimental background, such as the protein or cell under investigation, the conditions of its preparation, information about the imaging process and a record of the computer programs applied to the image information to reconstruct the final three-dimensional information. In cooperation with the Open Microscopy Environment developed by Prof Jason Swedlow and colleagues at the University of Dundee, the interface will be further developed to provide enhanced capabilities to its users. The Open Microscopy Environment is a set of software tools aimed at researchers in light microscopy to support the managemant of large image data sets. The collaboration between EMDB and OME for the first time links researchers in light and in electron microscopy to create improved tools for public research image databases, and leverages the technologies developed in OME to improve the general user experience. A particular goal here is to extend the user interface to accept 'rich submissions', i.e. complex three-dimensional image sets that contain a variety of information. This might include in particular details of the atomic structure of proteins derived by 'fitting' known atomic structures into the images, or annotations informing on the putative locations of molecules in larger structures. New computational techniques in electron microscopy also produce image sets showing structural variations that proteins undergo when performing in biochemical processes, and thus directly contribute to understanding the molecular machines that drive the processes of life. The technology developed in this resource will not only benefit biologists who use the resource, but also serves as a testing ground for future public biological image databases. As imaging technology in many areas develops, it is less and less often the case that a single image captures all the information a researcher needs, and more and more often large sets of images have to be used and communicated between researchers. Image databases and the new field of bioimage informatics will play a major role in this.

Technical Summary

The Electron Microscopy Data Bank (EMDB) is the central resource for storing and distributing three-dimensional density maps of macromolecular complexes and sub-cellular structures derived from electron microscopy. It has at this time about 600 entries, a large number of which are high-resolution molecular structures derived from single-particle analysis, in which many images are combined computationally to improve resolution, and a smaller but growing number from cryo-electron tomography, in which the specimen is imaged from different directions to give a three-dimensional reconstruction of sub-cellular structures and macromolecular complexes in their native context. Improvements in specimen preparation, imaging technology, and computational processing of the images are expected to increase further the volume and quality of images derived from these technologies. In particular, cryo-electron tomography is expected to yield images of the orchestration of macromolecular complexes as they perform the molecular processes of life in the cell, thus bridging structural and cellular biology, and single-particle analysis has now reached a resolution at which the molecular structure of proteins can be solved. The goal of this project is to extend the EMDB deposition and retrieval system to deal with the increased quality and volume of image information. A novel aspect is the collaboration with the Open Microscopy Environment (OME), one of the major software tools for managing image databases in light microscopy. Combining the experience of EMDB in public scientific image databases with technologies in OME aimed at user-friendly web-based interfaces, in particular its visualization server and its metadata management, will not only maintain EMDB's technical and scientific excellence , but also serve as an example for future image databases that will arise from the rapidly developing bioimaging technology.

Publications

10 25 50

publication icon
Eliceiri KW (2012) Biological imaging software tools. in Nature methods

publication icon
Linkert M (2010) Metadata matters: access to image data in the real world. in The Journal of cell biology

publication icon
Patwardhan A (2014) A 3D cellular context for the macromolecular world. in Nature structural & molecular biology

publication icon
Patwardhan A (2012) Data management challenges in three-dimensional EM. in Nature structural & molecular biology

publication icon
Swedlow JR (2009) Open source bioimage informatics for cell biology. in Trends in cell biology

publication icon
Swedlow JR (2012) Innovation in biological microscopy: current status and future directions. in BioEssays : news and reviews in molecular, cellular and developmental biology

 
Description we have substantially enhanced the production services available for the EMDB archive. These include EMSearch (pdbe.org/emsearch), a new search tool with a comprehensive range of search options, EMStats (pdbe.org/emstats), a data-mining tool with dynamic generation of statistical charts, a new 3D viewer (e.g., for EMD-2047: http://www.ebi.ac.uk/emdb-srv/atlas/2047_openastexviewer.html), and a new tool for the visual analysis of maps and map/model overlays (pdbe.org/emd-####/analysis where #### is the emd accession code, for instance 2047). Furthermore, the look and feel of the EM pages at PDBe (pdbe.org/emdb) has been completely revamped to bring it inline with the general theme of the PDBe website.

• We have built a full working prototype of EMDB running within the OMERO framework, imported the complete holdings of EMDB into this system, and incorporated a specially-adapted version of the OpenAstexViewer (OAV) software (a molecular structure and map viewer) into this system. This work required an extensive analysis of the memory footprint of the software, and modifications to minimise client-browser memory problems. We have incorporated third-party EM analysis and visualisation packages-EMAN (currently being developed by one of the US partners in EMDB, the laboratory of Prof. Wah Chiu in Houston, TX), Spider, and IMOD-into OMERO, demonstrating the incorporation of EM analysis tools into this project

• We have organised and held a community meeting-"Data-Management Challenges in 3D Electron Microscopy (DMCEM)" at Hinxton Hall, Wellcome Trust Genome Campus, Cambridge, UK on 5th and 6th December 2011. Participants included experts with established pipelines for 3DEM data collection and processing, and thus valuable expertise on the data-management challenges facing the field. Several participants have been involved in community-wide initiatives to define standards and conventions, and to define reporting standards for validation in 3DEM - the EM Validation Task Force (EM-VTF). Developers of several major EM software packages, including Appion, Bsoft, EMAN2, EMEN2, IMOD, MRC and Xmipp, also participated. Martyn Winn represented the CCP4 and CCP-EM projects. Representatives from this project-PDBe and OME also attended. The major output is a report defining future actions for EM data management and repositories. The report has been submitted for publication in Nature Methods as a Commentary.

• One of the DMCEM recommendations was the expansion of EMDB to include serial section and 3D tomograms. These will be hosted by an OMERO instance at EBI. This instance is installed, running and scheduled to be incorporated into the production EMDB by Q3 2012.

• The OMERO Image Viewer is being adopted by PDBe as a production slice viewer for tomograms in the EMDB archive.
Exploitation Route All the resources and technology developed in this project are available as open source software.
Sectors Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology

URL http://emdb.openmicroscopy.org.uk
 
Description Our technology has been incorporated into the PDBe, the definitive resource for biological structural data. It is used by 1000's of scientists every day.
First Year Of Impact 2012
Sector Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology
Impact Types Societal

 
Title EMPIAR Archive 
Description EMPIAR is the definitive repository for cellular electron microscopy and is now becoming a home for Volume EM as well. 
Type Of Material Database/Collection of data 
Year Produced 2015 
Provided To Others? Yes  
Impact With the technology we developed in this proposal, EMPIAR has been able to hold large, multi-dimensiaonl cellular EM data. 
URL https://www.ebi.ac.uk/pdbe/emdb/empiar/
 
Title OME Data Model Updates and Integration of OMERO into EMDB 
Description A major goal of this project is the incorporation of tools from the Open Mi-croscopy Environment (OME; http://openmicroscopy.org) into the EMDB re-source. Our efforts have focussed on adapting the OME Data Model to accommodate EM data and on the development of a customised version of OMERO tailor-made for EMDB. * The BioFormats library has now been extended to read several EM formats including MRC, Spider, Imagic, FEI, Gatan Digital Micrograph, JEOL, and IMOD. * We have built a full working prototype of EMDB running within the OMERO framework (EMDB-OMERO) and imported the complete holdings of EMDB into this system (http://cvs.openmicroscopy.org.uk/snapshots/movies/omero-4-2/mov/Extending-OMERO-EMDB.mov). * We have incorporated third-party EM analysis and visualisation packages (EMAN -from Prof. Wah Chiu in Houston, Texas-, Spider, and IMOD) into OMERO, demonstrating the incorporation of EM analysis tools into this project. A demonstration version of our work is available: http://emdb.openmicroscopy.org.uk * The same new version of the OpenAstexViewer offered as a web service for EMDB has been integrated into OMERO. * The slice viewer offered as a web service for EMDB tomograms uses the OMERO Image Viewer and has been adopted for the specific purpose by PDBe. To support this, Dundee has undertaken significant work to modular-ise the Image Viewer so that it can be used as a stand-alone component. With the EMDB-OMERO prototype and the slice viewer, we have demonstrated that we can read EM data into OMERO, edit meta-data using OMERO.editor, archive maps and meta data in OMERO and present data to external users. Crucially, EMDB-OMERO has the functionality to store and present image stacks, and to perform image processing using some of the commonly used packages in the field. 
Type Of Technology Software 
Year Produced 2011 
Open Source License? Yes  
Impact The software has been incorporated into a production system, the PDBe, running from EBI. 
URL http://openmicroscopy.org
 
Title The OMERO Platform 
Description OME Remote Objects (OMERO) is a modern client-server software platform for visualizing, managing, and annotating scientific image data. OMERO lets you import and archive your images, annotate and tag them, record your experimental protocols, and export images in a number of formats. It also allows you to collaborate with colleagues anywhere in the world by creating user groups with different permission levels. OMERO consists of a Java server, several Java client applications, as well as Python and C++ bindings and a Django-based web application. 
Type Of Technology Software 
Year Produced 2009 
Open Source License? Yes  
Impact OMERO powers a large number of public image data repositories, and is installed in >4,000 sites worldwide. Please note-- the year of output realisation is incorrect, as it is has been a development project running since 2005. 
URL https://www.openmicroscopy.org/site/support/omero5/users/index.html