Image Display | ||
DPS/Mosflm/CCP4 | ||
Denzo | ||
Mosflm | ||
HKL2000 | ||
Images once collected may be examined using the Adxv software on the DECalpha (bl96alpha2). This software is initialised with the command adxv which produces the following three windows.
Adxv Load: In this window it is possible to choose a directory, list the files within it and load an image.

Adxv Control: This window allows the user to alter the viewing parameters of the image. For example, it allows parts of the image to be magnified, the colour scheme of the image to be altered and the contrast of the spots to be changed, all for optimum viewing of the image. Clicking on File allows image files to be loaded or saved and the window to be exited. The View option enables a histogram of chosen peaks, the image header and statistics to be examined. In addition to this the Adxv Predictions window may be called up to search the image for peaks. The Edit option allows the beam center to be altered, the resultion rings to be switched on/off and background subtraction to be carried out.

Adxv: This final window displays the image. Clicking on the left/right hand buttons on the mouse while over this window enables areas of the image to be magnified and viewed in various ways and rows of spots to be chosen for lattice parameter determination. The resolution of a particular position may be obtained by placing the mouse at that position, the resolution and intensity are then shown in the upper left corner of the window.

If the crystal lattice parameters are unknown it may be possible to estimate some of them from an examination of a few images. Select a row of diffraction spots and zoom in on the area, this should cause a 'zoom' window to appear. Place the mouse at one end of the row of spots, hold down the left mouse button and drag the mouse along the row. A new window should now appear showing a plot of pixel values along the line and an estimate of the lattice spacing.
The DPS/Mosflm/CCP4 data processing software is initialised by typing process on the DECalpha workstation. It is recommended that a new subdirectory is created in which the data processing can be carried out. On initialisation the main data processing control window appears.

This window allows crystal and experimental setup information to be input and various processing parameters to be chosen in addition to controlling the data processing. Information relevant to the progress of data processing and some instructions are also relayed to the User via the white area at the bottom of this window.
The usual procedure to set things up for data processing would be as follows:
Please Note that if you have collected 2 x 2 binned data the software will have read this information from the data file header.
For data processing follow the flow chart across the centre of the main data processing window (as shown below).

The major steps in data processing using the DPS software will now be dealt with in more detail.
The indexing process involves visual inspection of one or more images, location of the diffraction spots and the determination of the unit cell parameters and orientation of the crystal that produced the images.
A single image may be indexed using either the Mosflm or DPS software. Clicking on Index: Mosflm causes the Mosflm window to appear, typing @index.inp then initialises the program and allows indexing to be carried out in the usual mosflm way.
For mosflm indexing it may be useful to use two images well separated in phi. Once the first image is displayed click on Find Spots. This should cause a small window to appear with some information about spot finding. A small red rectangular box should also appear superimposed on the image, this shows where data for a radial background correction will be taken from. If this box overlaps an unuseable area of the CCD change the Y Offset parameter and click on Find Spots again. A summary of the spots found should then appear in the window. Change the Threshold until the required number of spots is displayed. If another image is to be used for indexing click on Read Image and repeat Find Spots. Indexing may now be carried out by clicking on Autoindex. A new window should now appear, answering 'yes' to it's questions will cause the results of autoindexing to appear in yet another window. Choose the solution which has the highest symmetry but reasonable penalty. Type the number of the solution and a space group name. If the cell parameters then look ok answer 'yes' to update the cell parameters and accept the solution. Now use Predict to generate the diffraction pattern. Exit from Mosflm and save the new parameters in the main processing window.
Clicking on Index: DPS causes the DPS Autoindexing Control window shown below to appear.

First choose the image(s) you want to use for peak searching, then decide on the resolution limits and maximum cell edge. Clicking on Find Peaks initialises the Adxv software and Peak Search may be used to find the peaks (see here for details).
Once this has been done Autoindex should be clicked and a list of possible cell parameters (with their corresponding crystal systems and space groups) will appear. Each possibility has a number associated with it, the lower the number (usually under 100) the more likely the result is correct. The result can be displayed so that the fit of predicted peaks to observed peaks can be examined. Clicking on Display Result causes the Mosflm window to appear and typing @display.inp displays the image. It is possible to get an idea of the mosaicity of the crystal by examining the real and predicted spots along a lattice row. If extra spots are predicted the mosaicity being used is too high. If there are more real spots than predicted the mosaicity is then too low. Mosaicity may be defined in the Integration Parameters.
If everything looks ok the result can then be saved by clicking on Save Result (automatically updating the parameter file). Note that the result displayed and saved will be the one correspoding to the initial input information of Cell Parameters. Thus if a result other than this is desired the Cell Parameters must be altered and Autoindex rerun.
It is optional (though found to be not entirely necessary !) to refine the unit cell parameters and mosaicity in a preliminary run before integrating the whole set of images. Users may thus wish to ignore Refinement and go straight from Autoindexing to Integration of the entire dataset. Once autoindexing has been completed the unit cell may be refined using a block of images, thus the End Image No in the main data processing control window should be altered. Refinement of the distance, direct beam position and other camera constants is carried out automatically during integration. Before carrying out refinement the user may wish to alter some of the parameters as shown in the Integration Parameters window (see Integration section).
Clicking on Refine:Mosflm causes the Refinement control window to appear (as shown below). In this window it is possible to select some images for Mosflm to refine. Choose two sets of 2-4 images separated by at least 10 degrees in Phi. Decide whether to refine the mosaicity (ISOTROPIC, usually refined) or not (NONE). The current cell parameters can also be altered if desired. Refinement is started by clicking on Start Refinement.

The results of the refinement are then shown in the Refinement Control window. The message area in the main window will display the new matrix file, refined values for the unit cell, missetting angles, mosaicity and camera constants. If everything looks ok click on Save Unit Cell. If the direct beam position or distance have changed a lot then the new values may also be saved using Save Beam Pos, Dist. Save Everything will update all the parameters. If the changes in cell parameters or camera constants are substantial it may be better to do another run of refinement. After refinement you may want to hit Index:DPS to get the Autoindexing window and Display Result to check everything is ok.
Integration is carried out using Mosflm. The main parameters are set up via the main DPS window and the Integration Parameters window.

Other parameters default to reasonable values but they can be changed if the User desires. The important parameters are:
Image files to be processed - these are taken from the main DPS
window. Discontinous sequences of images may be integrated
separately.
Resolution limits - the outer limit is set in the main DPS
window. The inner limit is set to correspond to a 5mm radius about the direct
beam position.
Camera constants - beam position, crystal to detector
distance etc. are taken from the parameter file.
Crystal parameters -
symmetry is taken from the parameter file, unit cell and crystal orientation are
incorporated in the matrix file generated during indexing and updated during
preliminary refinement.
Beam divergence and Mosaicity - taken from the
parameter file.
Spot separation and profile fitting parameters -
Mosflm optimises the box used for fitting. Spot separation may need to be
changed in some cases (eg very large/small spots or spot elongated in one
dimension).
Detector parameters - gain, overload value and Add
Partials can be changed in the Integration Parameters
window if required.
Post refinement parameters - these are set using
the Integration Parameters window. If post refinement is enabled
the orientation of the crystal will be refined for each frame using the
information for intensities of partials on that frame and the next in the
series. Mosaicity and unit cell parameters may also be refined. However the unit
cell should not normally be refined during integration as it should have been
well determined by indexing and the preliminary refinement.
If you need to change the defaults for any other Mosflm parameters click on Edit Cmd Files Before Execution. Any parameter changes should be updated in the parameter file by clicking on Save. When integration is started a monitor window should appear to follow the output to the log file. When integration is complete a summary will appear in the main DPS window. This shows the refined values of the camera constants for each frame. For more detailed information the log file should be examined.
The output from integration is in the form of an mtz file which contains observations from each of the images in the data set. Sortmtz is run automatically after integration to sort the observations by Miller indices. After this the data may be scaled and merged using Scaling:Scala. The main parameters are set up via the main DPS window and the Scaling Parameters window.

Clicking on Start Scaling runs the program with it's default parameters. The names of files and the images to be used are obtained from the parameter file and are displayed in the main window. The default scaling parameters are as follows:
The default parameters may be changed in this window and in the Additional Parameters window. After scaling and merging have been completed click on I to F:Truncate to get the final files of F's in both binary (mtz) and ASCII formats. The programs used to do this are the CCP4 truncate and mzt2various.