Process the data with HKL2000
From X6AWiki
Experiment - HKL 2000
[edit]
Data Processing using Denzo
1. Organizing the folders prior to using Denzo
* Before doing anything , for each of your project, create 2 folders: process and images under /dataX/xyz/crystalID/denzo/ and in
process folder, create another folder Xfiles.
* Copy all the input files from /data3/Inputs/Denzo_Scalepack/ to your process subfolder.
2. Using Denzo
* Open 2 terminals being in process folder.
o First terminal will be used basically to view the image and examine .
xdisp-q ../images/image_001.img and click on peak search tab.
This will create a file called peaks.file in the same fodler. Keep this window open throughout the processing.
o In the second terminal you will be running Denzo.
First step will be to edit auto.dat file.
Run Denzo by typing denzo and type @auto.dat
It should use the spots found by xdisp and index your image.
If it's ok refine it.
3. Refining Steps
* In denzo prompt,
Type: fit x beam y beam go go go go go go go go go go go
(during this step you have to keep an eye on some parameters, chi2 error etc,.)
* Fit cassette rotx roty go go go go go go go go go go go
If the error is still big >> some more go
* Fit crossfire go go go go go go go go go go go go
* Fit crystal rotx roty rotz go go go go go go go go go go go
* Fit cell go go go go go go go go go go go
At this step you should increase or decrease the mosaicity that will depend on the number of spots predicted, if less -
increase it and if more - decrease it (just have a look on the xdisp window.)
You should also change the box size and spot size (radius or elliptical).
After each changes some go go go go go go go go
If it's better keep the new values otherwise keep the old one.
* Fit all, fix distance go go go go go go go go go go go go go go go go go go
By typing list you have all the refined parameters distance etc ...
4. Integrate your dataset
In denzo window type @integ.dat
During the integration keep an eye on the denzo window and xdisp window. For any problems during the integration (For example: less
predicted spots, decay between predicted and real pattern etc..), start again from scratch by re-editing auto.dat doing changes to
other parameters like unit cell, space group, beam position etc,.
5. Scaling
* You need to edit the file scale.com
In the first run uncomment @reject in the scale.com file.
* Run scalepack
scalepack < scale.com > scale.log
* Check from the log file, the new refined parameters, put them in scale.com file and re-run the scalepack.
The above steps need to be run several times until the results are satisfactory.
* Last step is to comment @reject in scale.com file and re-run scalepack.
NOTE: For SHELX, CNS, ccp4, you need to unmerge the original indexes. For more help, go through the HKL2000 manual at the X6A beam line or visit http://www.hkl-xray.com
