The current minor release version is 1.1.0.

A number of movies is available that demonstrate installation, setup, and running FOCUS. The full playlist can be found on YouTube

Introductory and Tutorial Movies

FOCUS requires C++, gfortran, Qt, tcsh and Python. It uses several third-party programs in the background. Several of these don't run under Microsoft Windows. For these reasons, FOCUS only works on LINUX or OS X machines. If you plan to use MotionCor2 or gCTF or other NVIDIA-GPU-enabled programs in the background, then you have to stay with LINUX, since Apple doesn't support NVIDIA GPUs unfortunately. However, if you plan to work with 2D crystals or only third-party programs that don't require NVIDIA GPUs, then OS X is also a suitable platform, provided you have at least 8 GB RAM.

Hardware recommendation for Linux

Hardware recommendation for Mac

Focus depends on Qt5 for deploying a GUI. Other EM Suites are used wherever required. Find information on how to download, install and setup Focus software for the first time.

Download the software

Installation from source code

Setting up external software tools

Default installation directories

Once Focus is installed it can be easily launched using the Command line or Desktop utilities provided with Focus. Following links give more information on launching Focus, adding a project, configuring global user settings and brief introduction of the GUI.

Launching Focus

Adding a new project

Introduction to GUI

Configuring user settings

The links below would help you to start using Focus. The first things to do in any project is to import images and later once imported process them and manage them. Processing depends on the type of EM method used and is detailed in later sections.

Import images and movies

Managing Images using Project Library

Focus allows to process images via the precompiled scripts. Each script calls a few executables (external or internal) and fetches parameters associated with the project or the image from the configuration files as required. Processing can be either done manually for each image allowing to tune-in the parameters and later a bunch of images can be processed in parallel.

Manually processing an image

Parallel Processing

Remote monitoring allows to see the status of the processing in Focus from a website that can be accessed remotely. This in turn helps to check the recording status of the microscope.

Setting up the server

Configuring Focus for remote monitoring

Collecting data in Focus

Setting up apps on Android and iOS

Focus includes scripts prebuilt to carry out complete 2D crystallographic image processing. Following pages describe how to start from images/movies, create final 2D maps, merge the data to produce 3D and finally refine the reconstruction to minimise the effects of missing cone.

Image Processing of 2D Crystal Images

2D Merging

3D Merging

Refining Merged Map

2D Crystal Test Images

Once the images have been processed they can be exported. The export function in Focus organizes the image files and metadata in the arrangement required by the target package. For example, if the target package is RELION, the data will be reorganized so that all aligned average images and a single metadata STAR file are created in a single subfolder. On the other hand, if the target package is IMOD, the electron tomography data will be reorganized so that there is one MRC stack for each specimen location, containing the entire tilt series ordered by tilt angle, irrespective of the recording order (e.g., if they were acquired using the Hagen scheme).

A list of FAQ is available in the following page:

Frequently Asked Questions

From 2006 through 2016, we organised workshops aimed at teaching electron crystallography of membrane proteins

The fifth edition of our hands-on workshop on “2D Electron Crystallography” (hosted at C-CINA)

Workshop 2016, C-CINA

Workshops hosted at UC Davis and at C-CINA

Electron Crystallography Workshops Workshop 2012, UC Davis

Workshop on Electron Crystallography, hosted at C-CINA.

Workshop 2010, C-CINA

Workshops hosted at UC Davis and at C-CINA

Electron Crystallography Workshops Workshop 2008, UC Davis

Workshop hosted at UC Davis on Electron Crystallography.

Workshop 2006, UC Davis

A pipeline of the programs, files and data involved in the processing of 2D crystal images is available here

Find documentation of some of the important parameters involved in processing of 2D crystal images with Focus here

A list of frequently asked questions is complied and can be accessed here. Please use this list to find your question, and if you do not find any help, please contact us here.

Video tutorials are available here. Also refer to the workshops for more details.

Explanation of the 2dx GUI

read more...

Beginning an Image Processing Project

read more...

Status Pane

read more...

Processing an individual image

read more...

The Full-Screen Image Navigator

read more...

Refining a lattice

read more...

Refining the CTF

read more...

Spot Selection

read more...

Tilt Geometry Definition

read more...

ML algorithm applied to 2d crystals

read more...

Data Flow

These pages describe the flow of data through the different programs. File names, formats and contents is listed. 2dx and the underlying MRC programs process and then evaluate images, and extract values for amplitudes and phases of Fourier spots. These are then handed over from one program to the next one, where each program does something to those values (e.g. correcting for the CTF, or merging several values together). Each program thereby unfortunately needs a different type of data that is therefore often in a different format, and the whole thing quickly gets very confusing. Most of these files are text files (ASCII), though, so that you can always open them with a text browser (vi, edit, SimpleText, or even Word (but don't save it back as “Word” file.doc file). We have here compiled a list of all the programs that are called in sequential order, and have listed the input and output files for these programs. For the output files we have then included a short description of the file format, followed by an example of how that should look like. This is only for reference purposes, in case somebody tries to follow the flow of data from one program to the next one. read more...

There are currently no items in this folder.

changes.htm

######0.9.1-7ef0010###### Change Logs

0.9.0 (Developmental Pre-Release)

• First implementation of the Focus package.
• Contains features like import tool, parallel processing tool.
• Various modes can be set depending on the type of the EMproject.

Old 2DX Change Logs

• 4.0.0 October 2016
• 2DX has a new integrated Graphical User Interface called 2dx_gui
• Both merging and image processing can be done with 2dx_gui
• 2dx_merge and 2dx_image are now just command line
• 2DX Projects can now be ported to single particle packages like FREALIGN/RELION
• ZORRO integrated
• MTZ write capabilities added
• FSC/Phase residuals added to the Quality Evaluation script
• Features added to the core kernel
• [bug] Major bug removed in parallel refinement script
• Fixes in movie mode processing

3.5.0 Dec. 14, 2015

• A new tool to automatically switch between maps and references now available
• Parallel processing of images in 2dx_merge is now possible
• Weighted Backprojection introduced instead of Latline fitting
• 2D merge and 3D merge separated
• Movie mode processing for direct electron detectors
• GUI Improvements in 2dx_merge and 2dx_image
• Moved to Qt5.x
• Improved build and packaging, no dependency on Boost
• It is now possible to sharpen maps using bead models
• B-factor sharpening scripts included
• New Volume refinement methods included which helps to encounter problems of Missing cone
• New kernel tools introduced, capable of doing all the major operations on EM-Volume
• Radiation damage measurement for movie-mode processing
• [bug] 2D-ML now works with the newly integrated general processing parameters
• [bug] Classical diffmap-script working for maps with where RES < 6A
• ionalityMany other bugfixes

3.4.2 Feb. 26, 2014

• Full OSX-10.8 and OSX-10.9 support
• Auto-saving added
• Move image to another project now supported on right-click, useful in case of two crystal conformations
• New bug tracker and developer wiki
• TTF correction script added for images with a higher tilt then a given value
• Internal version control system switched to git
• ttfcorfitst removed from Unbend2, as we do not recommend this option at all
• dx_automator added
• Automated drift-correction GUI added
• Single Particle Processing scripts added, but not ready to use so far
• Difference-map functions added
• Command line merger
• [bug] Beam-tilt refinement now working
• [bug] Maximal index in CCP4-script changed in order to achieve high resolution (better than 6A)
• Latest version of CTFFIND3 integrated (incl. OMP)
• Refine switch now also refines the tilt-geometry
• [bug] Resolution plot fixed
• [bug] Error on PLTILTK fixed
• [bug] Bug in astigmatic handling of CTF fixed
• File initialization improved
• [bug] Synchronization for .raw.mrc files fixed
• [bug] Reference origin resetting works now
• Parameters are reset for a new image now
• Optimal image import for K2 Summit data
• Selection of multiple images with right-click now possible
• Possibility to generate non-CTF-corrected map
• Removed re-unbending from the standard script list
• Added options for tilt-angle range statistics
• Included fomstat in 2dx_merge_merge again
• Masking both lattices at the same time now possible
• [bug] Issue with symmetry p4 in hklsym fixed
• PDF translator added
• Better sync with backup script
• NANs removed from hklsym4-output
• mtzdunp now displays all reflections
• Non-scroll QComboBox added
• refine Defocus now uses only spots up to IQ-4
• FOM calculation for symmetry-constraint lattice lines implemented into 2dx_latlinek.for
• Lattice line in color now
• Tiff translator added
• Better spot scan masking functionally
• Support for TVIPS-headers
• Origtilt output extended
• 2dx_hklsym3 removed
• [bug] Contour plot in p2 now working
• Amplitude scaling added
• Quality evaluation added (2D and 3D)
• Longer image names now supported
• MergeIQMAX added
• [bug] Unsigned/singed integer issue finally fixed
• [bug] TLTPLOT now used correctly
• Second lattice determination
• Shortcuts to image/merge directory added in 2dx_image/2dx_merge
• Shortcut to logbrowser added
• Updated change_dylib.sh
• More stable Linux support
• SBGrid integration
• Full build-process done in cmake now
• Compilation optimized
• SCALIMAMP3d added
• Final map exported also as .tif
• Sub-volume generation of final map
• Use of cpack for Linux package generation
• Improved trash functionality

3.4.1 Apr. 9, 2012

• PSF in 3D now calculated
• tilt geometry variable naming simplified
• imagenumber is now also updated, when a default config is used with a wrong image name and number and a new MRC image is found
• 8×8 binned downsampled nonmasked image now calculated by FFTIR script, for visual tilt geometry from Thon ring fitting.
• Changed MRC programs from INTEGER*2 to INTEGER*4
• Tilt geometry refinement from MTZ lattice lines can now be done with GUI-definable parameters
• Extended reference maps are displayed in 2dx_merge for each image
• Refinement now puts out reference APH files
• spotlist is copied from masked to nonmasked name, if needed
• allows masking multiple times and with as many vertices as 2dx_boximage can handle
• the synchronized property is now also changed in the upper 2dx_master.cfg and we have two save functions, one that synchronizes with the upper config and the other not.
• you can now change the threshold for the background and the step size for the contours in the CCP4 map generation
• when a parameter in 2dx with the attribute SYNC_WITH_UPPER=yes is saved, it is now also stored in the upper config.
• getLattice now favors the vectors that span the smaller angle when reducing them to the shortest even if the one vector is bigger by a delta then the reduced one
• allspace now picks the highest symmetry, whose residual is still below the target (=acceptable)
• [bug]fixed bug in file selection in the Synchronize Project with Backup script
• have the C Shortcut to toggle the thon rings back
• Lattice line plot output updated
• TAXA display in 2dx_image and 2dx_merge now correctly respects revhk, rot90, etc.
• Backup/Restore Databases script created.
• CTFTool now allows toggeling of the Thon ring overlay
• the Lattice Refine Dialog now shows the selection of a point
• changed internal structure of the albumModel to enable more features in the imageAlbum
• [bug]fixed bug in display of sheared final maps in Protein Album
• Automatic masking in Unbend-II accelerated.
• generate_sym_map now works also in p2. A header line was ignored in 2dx_permutate, now fixed.
• [bug]bug in 2dx_image script generate-sym-map fixed
• Cloning of Image Directory updated.
• Synchronization Script created
• raw TIFF image handling cleaned up.
• FSC-weighted envelope function implemented into Maximum Likelihood code.
• cleanup leaves LOGS/*.results in place
• ML output map increased, envelope convolution activated.
• Script to Repair directory structure added.
• Implemented verification of correctly linked 2dx_master.cfg files.
• the getLattice script now also allows to refine the tiltgeometry with emtlilt without having to determine the lattice automatically
• [bug] SCRATCH_DISK and FFTIR_DISK work even if the original directories do not exist
• Inverse Fourier Filtering implemented
• color output in distortion plots
• [bug]synthetic reference script debugged
• only asymmetric unit produced during merging
• script to delete images (move to TRASH) updated
• 2dx_hklsym.for now can do p4212
• added parameter refinement script for synthetic unbending
• added custom scripts to refine parameters for Unbend I & II first in a coarse grid search over box & mask and then a fine refinement
• ctffind3 now works on nonmaskimage, instead of image
• Refine Defocus from merged data script
• 2dx_refine_unbend2 now also allows the refinement of maskb02
• 2dx_makedirs checks if dirs are empty or not before linking and moves the existing files
• 2dx_makedirs now creates the symbolic links to the SCRATCH and FFTIR disk
• QVAL table is now written as a result of the simultaneous refinement of boxb/boxa and maska/maskb for the Unbending
• 2dx_image.cfg is now updated appropriately: new paramters are added, changes in parmater properties are updated, and user set parameter properties are kept
• SCRATCH_DISK and FFTIR_DISK implemented.
• logbrowser becomes the acitve window when popping out of 2dx_image
• ztarwin warning only when modus=3D
• New Script added to clone an image directory
• updates in script Mask Crystal from Polygon
• initialization moved into separate script
• 2dx_maketran.for altered, to prevent gfortran compiler crash
• shearing of the generated CCP4 2d map is corrected
• [bug]fixed bug #32: update version text. Plus the http get request is now done the qt4 way.
• verbose logging of merging statistics in 2dx_origtiltk implemented. An additional file 2dx_origtiltk-reflections.log is created now.
• [bug]bug in 2dx_latlinek resolved, where ZSTG was not initialized (reported by William Nicholson)
• progressStamps are visible again
• changes to get the resources path
• IS_WRONG is now depicted by the background color
• Automatic Scaling to INTEGER*2 0…16000 implemented
• Now also reads *.tiff in addition to *.tif
• [bug]bug removed that led to crash when 2dx_image.cfg was outdated.
• Delete Image as script added.
• PLTILTK in 2dx_merge included.

3.3.0 Dec. 24, 2010

• [Bug] 3D reference values from lattice lines are now correctly sorted, so that the creation of synthetic unbending references via MAKETRAN now works.
• [Bug] Fixed a bug in synthetic unbending when using TTF.
• Added a script option in 2dx_image for the tilt geometry refinement from defocus gradient.
• 2dx_merge now plots FOM-Symbol plot for non-tilted merged data.
• 2dx_merge now creates correct symbolic links for the configuration files 2dx_master.cfg upon initialization of a new project. These links are needed for correct synchronization of global parameters (e.g. synchronization of the crystal symmetry).
• Added the configuration file 2dx_merge-user.cfg into the user's home directory/.2dx, which stores user settings for 2dx_merge. It is also used by the file browser to navigate to the last used project directory.
• [Bug] 2dx_image becomes now the active window even when its called from 2dx_merge.
• CCP4 can now also be located in /Applications on MAC OS X.
• The local database 2dx_merge.cfg is now updated, when new parameters are added to the master configuration file /usr/local/2dx/kernel/config/2dx_master.cfg.
• Added a new parameter attribute ISWRONG, which when set to YES visualizes the specified values in the GUI in grey. This depicts that these values should be altered before processing. By editing them they become black on white, and ISWRONG is set to NO. This allows to have parameters prefilled with default values that are technically functional, but still wrong. For example, the default setting for defocus is 5000.0, but the exact value still has to determined. The defocus entry of 5000.0 is therefore initially labeled as ISWRONG.
• The variable attribute GLOBAL in the CShell scripst has been replaced with DISPLAY to make its functionality more clear. Scripts containing the attribute GLOBAL will still run.
• Updated the merging scripts so that Maximum Likelihood (ML) processing results from 2dx_image can also be used for merging. For this purpose, a new parameter ML_use_for_merging has been introduced, which allows to decide, if a ML result should be used for merging, or if the Unbending result should instead from this image be used for merging.
• [Bug] Fixed a bug in the ML code, which resulted in the Amplitude and Phase data to be written out with 180deg phase shift with respect to the Unbending results.
• Added a feature that checks if the symmetry parameter setting for ML processing is following the same space group symmetry as otherwise choosen for the project.
• [Bug] Fixed a bug in the ML code, which resulted in the Amplitude and Phase data to be written out without the initial image idenfier number. This prevented this file from being able to be included in the merging.
• Added a parameter mask_reset_reflocation that allows to decide, if during masking and optionally cropping of the crystal area the reference location should be reset to the new center of the crystal, or if it should remain in the old position.

—

2dx_gui_userguide.pdf

2dx_manual.pdf

… this page is under construction …