Collaborative Computational Project
|Home||About CCP-EM||CCP-EM Projects||Downloads||Resources / Documentation||Workshops / Courses||Positions|
Due to our close association with the CCP4 project, we have a strong interest in maintaining this format. Standardisation of the format is important, so that e.g. crystallographic map manipulation tools can be used on EM maps.
We believe that there should be no differences between the standard MRC and CCP4 formats, and any incompatibilities found are bugs! Known issues are:
More generally, a number of extensions are already generated by existing EM software. The question is which can be included in the MRC standard, without breaking other tools. We have agreed an update to the MRC/CCP4 format, called MRC2014. Details of the changes have been published in Cheng et al. (2015) "MRC2014: Extensions to the MRC format header for electron cryo-microscopy and tomography" J. Struct. Biol. in press. See separate page for further details.
FindEM is a series of programs, written mainly in Fortran, developed by Alan Roseman (University of Manchester) to help with automated particle picking from electron micrographs. There are five basic steps to the process, which include an optional initial reduction in sampling and Fourier low-pass filtering of the micrographs, running a program to calculate correlations between a previously determined template of a structure and the micrograph, and using a graphical-user interface to modify the parameters used to select the particles using the correlation functions, so that a user can finely tune the particles that are picked by the algorithm.
Currently, it is only possible to run the FindEM programs on Linux using c-shell. One of the early CCPEM projects is to make FindEM more user-friendly, by creating installation files for the FindEM programs and all dependencies (e.g. the MRC Image Library, which is used extensively within FindEM), streamlining the process by which all particles are picked, and enabling cross-platform compatiblity, by rewriting Unix-specific code in python, enabling them to also be run on Windows. In conjunction with Alan, we will also increase functionality, based on users' requirements, offered by FindEM.
Paper is out: Farabella et al., J. Appl. Cryst. 48, 1314 - 1323 (2015). [doi:10.1107/S1600576715010092].
One of the current challenges in structural biology is how to accurately determine the structure of macro-molecular assemblies. Using EM maps in conjunction with atomic structures of assembly components derived by other means (e.g. crystallography, NMR, or protein structure prediction approaches) can often help piece together the structure of the assembly. Although this fitting can be approximated manually, it is preferable to use automated methods in order to improve accuracy and efficiency. A number of metrics can then be used to determine the most energetically-favourable configuration of the components.
TEMPY is a Python library designed to help the user manipulate and analyse atomic structures and density maps from 3D EM. Based on the Numpy and Scipy packages, TEMPY will eventually be expanded to include a suite of functions for fitting and the assessment of fits, as well as being a toolkit for users to build their own, personalised, functions.
It is currently being developed by Maya Topf's group at the Institute of Structural and Molecular Biology (Birkbeck/UCL). CCPEM aims to implement TEMPY into a more general EM software framework, so that users can take advantage of the powerful modelling approach through a graphical front-end. It is hoped that by linking TEMPY with other parts of the EM pipeline, we will make it easier for (primarily lab-based) biologists to determine structures at the macro-molecular level with a smaller learning curve than is currently required.
Since 1st January 2015, CCPEM has taken responsibility for the maintenance and distribution of the MRC image processing software package, see Jude Short's message on the 3dem mailing list. This includes the 2D crystal and helical programs, the display program Ximdisp and many other utility programs. The package is available from the Downloads page.
The 2dx project maintains a "slightly adapted" version of the MRC software code, which is useful for 2D electron crystallography.