Chapter 1: Starting the Program and Setting the Data Sets


Step 6: Experimental details


Step 6: Experimental details

Check the parameters under Experiment Geometry and Frame Geometry to make sure they are correct. Having the correct x beam and y beam values is critical. Other parameters of the experimental geometry are not as important and mistakes can be found during the initial refinement. For instance, if the Oscillation Range is wrong, you will see that increasing the 3D Window does not increase the number of refined spots and instead results in an increase in the χ2. In that case one should move to 3D Window equal 1, process the data and observe Crystal Rotation diagnostic window. The most common cause is the opposite rotation of the spindle axis that can be corrected by correction of the goniostat definition in site window.

What if I selected a set by mistake and want to remove it?

That’s easy! In the main window, click the select button of the set you want to remove. Then click the remove set button on the right middle side of the main window (Figure 12). The program will ask if you really want to do this, and if so, the set will be removed.

Figure 12. The Set Control window

What do the Collect, Denzo, and Scale buttons in each window do?

This tells the program what you want to do with the data in this set. If you want to collect the data (collect), index and integrate the frames (integrate), and scale the data (scale), you would click all of these buttons (Figure 13). The default is for all the available options to be selected when the set panel is displayed (HKL-2000 control of data collection is still under development and currently supports only data collected on SBC beamline at APS and also on a few selected detectors). This is most useful if you want to jump to the end and scale frames that have already been indexed. In that case, just un-click the  denzo button but leave the scale button clicked, and the program will know that you are going straight to Scalepack.


Figure 13. The Select window options

What if I don’t want to process all of the frames in this set or I want to process some of them separately?

Notice that under the panel called Set Controls there is an option called split set (Figure 12). This splits your data set into two or more groups, which can then be integrated consecutively, independently, or selectively. Why would you want to do this? One reason could be if you had a bad or blank frame somewhere in the series of frames. Another possibility is that you changed the wavelength in the middle of the data collection. In general, sets are split whenever there is an interruption in the contiguous series of data collection. To split a set, select the desired set and then click on the split set button in the Set Controls panel. A new box will appear which has two sliders (Figure 14). Move the top slider to define the first set, and then move the bottom slider to define the second set. If you want to skip one or more frames, just make sure that they are in the “gap” between the two sets. When done, click on the split button and a new set will appear (Figure 15). Make sure that the Distance and Oscillation Start values are correct for each set (they should be read from the header of the data files, but you never know).

* In principle you can split a data set as many times as you need to, however, keep in mind that 3D integration cannot work across the boundaries between split sets.


Figure 14. The Split Set window

Figure 15. Data set divided into two sets with removal of four images in the middle

The Distance that came up in the Experiment Geometry window is wrong. How do I fix it?

At the bottom of the page you have obviously seen the boxes labeled Experiment Geometry, Frame Geometry, and Exposure Time (Figure 16). For processing, the most important things you will have to fill in, are the Distance and the Frame Width. These values, as well as values mentioned below, can be changed in the table under the Summary tab (Figure 25) or/and the Edit Set(s) window (Figure 9). Distance may be incorrect because the header of many data files does not contain this information, or the information in the header itself is incorrect. Frame Width is more likely to be right. You should also make sure that the Wavelength is ok, too. For home data collection setups the wavelength is usually defined in the Site Information file, but obviously for synchrotrons the wavelength will vary for each experiment. You can fill in the correct values for the Exposure Time and Oscillation Start and so on, although these won't affect the results from processing a single contiguous data set. However, if you are processing more than one data set with a shared set of parameters (like crystal orientation, etc.) Oscillation Start will be important. When in doubt, put assumed values and try manual refinement. If you see that Distance, Width, or Wavelength are consistently wrong contact the beamline or detector vendor and ask for header correction.

Frame Geometry: It looks like I can put in more frames than I actually have. What is this for?

This is particularly useful if you are processing data as it is being collected and you want the program to integrate the frames in the same session, but you don’t want to wait to get started. In the Edit Set(s) window (Figure 9), enter the correct Number of Frames you are planning to process in this batch. You will notice that as you enter this number the Number of Frames value will increase in the Frame Geometry window, as well as the Exp. End in select panel at the top of the page. Notice as well that the End value in these two fields does not change; this is the number of frames that the program knows is on hand at the moment. For example, suppose you are collecting 100 frames of data, and you have 20 collected so far. You would enter 100 in the Number of Frames box before you started the integration of the 20 frames collected so far. You could then start the integration job and the first 20 frames will be integrated. When the last frame on hand is processed, the program will wait – giving the message “Waiting for Frames” – and when the new frames are written to disk the program will continue to process them. This is especially helpful at places like synchrotrons where the frame integration rate is a substantial fraction of the data collection rate and you want to maximize your efficiency.

*  Another way you can use the Number of Frames value is to stop the integration before all the frames are processed. This is helpful if you have a lot of frames that have lots of spots and so take a long time to integrate. Perhaps you just want to process a wedge of this data and get some idea of its quality. For example, say you have 300 frames but just want to look at the first 30. Enter 30 in the Number of Frames box and start processing. The first 30 frames will be integrated and then the program will stop. This is easier than splitting the data set and more civilized and nicer than hitting the abort integration button.


Figure 16. The experimental details panel


Setting up data files


Table of Contents


Display Your Image