Manual

Editing Absorbance Equilibrium Data:

All experimental equilibrium data aquired on the XL-A requires substantial editing and preliminary diagnostics before the data can be successfully analyzed with various analysis methods. UltraScan assists you in this process by handling most essential diagnostics and editing steps automatically, but permits you to intervene, where necessary. Editing is started by selecting
 

from the main menu of UltraScan.  This will bring up the editing window with the plot area and the control panel.

At the bottom of the control panel is a "Wizard" that provides step-by-step instructions to guide you through the multi-step process of equilibrium data editing. Refer to it to determine which action is required next.

Step 1: Select a data directory that contains the data files acquired by the Beckman data acquisition program. Such a directory can generally be found as a subdirectory of the Beckman data acquisition software. The default installation for the Beckman software is C:\XLAWIN (under Windows). Under Unix, the drive with Windows 3.11/95 will generally be mounted under the /dosc or /dos directory.
The experimental data will then be located in a subdirectory of the form mmddyy/hhmmss (month/day/year/hour/min/sec), identifying the date and time on which data acquisition commenced for a particular run. A file dialog will allow you to specify the desired directory. If the directory contains files of the type 00001.ra1, 00002.ra1, ..., etc., then it is a proper data directory. Depending on your method of transferring files from the Microsoft Windows drives to the Unix machine, your filenames may be upper- or lowercase. Make sure the filter is set to the proper setting in the configuration panel. If not, the scans will not appear in the dialog box and the directory will appear empty. Please note that you should have at least 3 scan files for each cell and wavelength to perform a half-way reliable equilibrium data analysis. Once you have navigated through the directory tree to the desired data directory, simply press the OK button. Next, the program will load the data in this directory and determine a number of important diagnostics from the scan files present in this directory:

1. The number and position of cells that were scanned in this run. If the sequence of the scan files is out of order, a warning message will be shown. If this message shows up, and you acquired the data at a single wavelengths, you can rescue your data by rebuilding the data sequence with the Scanfile Reordering Utility
2. The number of wavelengths and their settings that each cell was scanned with
3. The number of scans that were taken for each cell and each wavelength
4. The average temperature over the entire run
5. The temperature of each individual scan
6. The program will determine if the temperature varied for more than a pre-set tolerance value over the course of the run and warn the user if the tolerance was exceeded. In such a case, a warning message is printed forcing the user to acknowledge the diagnostic before editing can resume. Depending on the method of analysis of equilibrium runs, a temperature variation during equilibrium experiments may actually be of little consequence, since the software will compensate automatically for variations in temperature-dependent density changes. When measuring thermodynamic parameters such as association constants it is important to realize that temperature variations may significantly affect the results and should be avoided if possible. Also, for velocity experiments a constant temperature is required throughout the experiment in order to avoid systematic errors introduced through density and viscosity changes, as well as variations in the thermodynamic properties of the sample.
7. The length (duration) of the entire run in hours and minutes
8. The rotor speed at which each measurement was made
9. The identifications for each cell that were entered at the begin of the run.

The purpose of this window is to facilitate the identification of experiments from the rather cryptic information provided by the file- and directory names created by the Beckman data acquisition software. Information about various cells available for a particular run can be obtained by selecting the appropriate cells in the listbox (UltraScan allows for up to 7 cells to accommodate experiments performed with the 8-hole rotor AN-50 Ti, and allows for an unlimited number of scans to be analyzed for each dataset. The cell description will be updated for each cell that is selected in the listbox. If the description for a particular cell was not sufficient, it can now be updated and edited to accommodate changes.

At the lower left corner of the diagnostics window you will find a list of speeds for this run. If you want to know how many scans and wavelengths were collected for a particular speed, you can click on any of the speeds to obtain a window with the listing for that speed.

Step 2: Before editing of the data can proceed, a unique run identification needs to be entered in the box labeled "Run Identification". This run identification should not contain any spaces and should help the user in identifying the run by name. Spaces can be replaced by underscores. Please note that unlike under Microsoft Windows, file names will follow Unix name conventions (case sensitive). For example, a run identification such as "Chromatin_pH8" is distinct from "chromatin_ph8". A practical way of naming runs is to use the logbook number of the run, and to append a "v" for equilibrium-, or an "e" for equilibrium runs. If you change your mind and do not want to edit this data, you can cancel the editing process at this point by clicking on "Cancel" and you can then select a different directory instead. However, if you do want to proceed, click on "Accept". If you forgot to enter a unique run identification, you will be reminded by a modal dialog.

Step 3: Verify if the cell description is correct and adequate to identify the contents of each cell that contains data. If not, you can now edit the description for each cell by modifying the string in the editable text box. By changing the information in the "Cell Description" field, it is possible to include a more descriptive cell identification for the sample than was chosen at the time of data acquisition. After entering a unique run identification, the window will close and you will return to the main editing window. The control panel will now show all fields updated with the proper information obtained during the diagnostics.

Next, make sure the setting for the number of channels agrees with the centerpiece used in the experiment. 6-channel centerpieces are useful only for equilibrium runs and should not be used for velocity experiments. The checkbox for the default setting of 2 channels/centerpiece (regular 2 sector cell) should show the "checked" indicator at this point.

The number of editable datasets will be displayed in the "Run Detail" box. A dataset is the collection of scans acquired for a particular wavelength and a single cell. For example, if the cell in position 3 has been measured at 260 nm, all scans measured at this wavelength for cell 3 are considered to be a single dataset. To load the first dataset for editing, click on "Start Editing". All scans will then load into the editing plot window, and the status bar indicates the progress of data loading. Double-check to make sure the centerpiece, channel and rotor settings are correctly selected before proceeding. If the dataset corresponds to a 6-channel centerpiece, select the 6-channel centerpiece checkbox now by clicking on it.

Step 4: After loading is completed, the data will be displayed in the editing window. If you used a 6-channel centerpiece to collect your data, make sure to click on the 6-channel centerpiece selection button. Also verify that the rotor and centerpiece material (epon or aluminum) are correctly selected. If not, click on the appropriate button before proceeding.

Step 5: Now you need to define the meniscus limits. Those are the limits between which the data contains the meniscus for all displayed scans. These menisci may differ, because of rotor stretching, which causes the meniscus to shift at different speeds. The program will assign a unique meniscus for each scan by searching for the highest absorbance of each scan between the limits you have entered. First, click with the mouse on the left limit, then on the right.

Step 6: In the next step, you have to defined the limits for the useful data range for all scans. This range defines the data used in the data analysis. This region should exclude the meniscus and the bottom of the cell. First, the left limit is defined by dragging the mouse to a position near the right of the meniscus, and then clicking on it. If in doubt, rather include a little more data, it can be excluded later in the global analysis by fine tuning the editing limits.

Step 7: Next, the right limit is chosen in a similar fashion. The data will be rescaled to the new limits, and the extraneous datapoints will be discarded. If you have data from a 6-channel centerpiece, make sure to click on the 6-channel button. This will cause the program to prompt you three times for the above information, once for each channel.

Step 8 (optional): Now you will have the opportunity to exclude and edit scans. To edit the spike in scan 9 from this example, click on the ">" - button until it reads "9". The displayed scan will be highlighted in magenta to assist you in finding the appropriate scan (if you click on the ">>" and "<<" buttons, you can move around in steps of 10, which may be useful if you have a large number of scans in your dataset). Once you found the appropriate scan, click on "Edit Single Scan". Clicking on "Edit Single Scan" will bring up a window showing the selected scan by itself. Use the mouse to drag a box around the problem area. This process can be repeated until the desired zoom level is attained. Note: the program will automatically autoscale the y-axis, therefore it is not necessary to include everything in the y-direction. Only the x-limits are relevant. Once the desired zoom level is reached, the zoom function has to be turned off by clicking on "Turn Zoom Off".

Step 9:  Then you can click with the mouse near the problem point and it will be highlighted with red crosshairs. Next, simply click on the point where the point should be. You can click as many times as you want until you get it right. Note: Each time you move the point, the plot area will be autoscaled to best distribute the points on the available screensize, which may lead to a somewhat "jagged" appearance of the points. Once they are re-scaled to their proper dimension, they will appear normally again. Once you have found the optimal position for the point, click on "Accept". This will return you to the main editing screen.

Step 10 (optional): Next, you may want to exclude a number of scans from the run. Note: you can leave all scans in the dataset now and only select desired scans during the analysis. If you want to exclude scans now, you can accomplish this by clicking on "Exclude Single Scan" or a range of scans by clicking on "Exclude Scan Range", after setting the scan number in the counter. ">>" and "<<" buttons move in steps of 10, buttons ">" and "<" move in single step mode. As you select scans, the about to be excluded scans will be highlighted in red. After clicking on one of the "Exclude..." buttons, the highlighted scans will be deleted from the dataset.

Step 11: Once you are satisfied with the subrange of edited data you can save the subset by clicking on the "Save" button.
 
Now the data for the first dataset is edited and the program will automatically cycle to the next dataset by returning to the main editing window and updating the progress bar to indicate the status of file loading. Repeat Steps 4 - 9 for all datasets in the run. If you have selected the 6-channel centerpiece setting, the you will be able to select the next data range for the same cell and wavelength before the program cycles to the next dataset. This way in effect you create 3 sub-datasets for each dataset, for each channel one. When all datasets are edited, the program will return with a message window indicating that all scans have been edited and the data has been written to a binary file.

Last Step (optional): In order to conserve diskspace and to backup your experimental data, you can also archive your raw data into a compressed unix tape archive by clicking on "Archive Raw Data". This will create an archive which is a single file containing all scans acquired by the Beckman Data Acquisition software. Note: This function will not remove the original files from your computer. This step should be performed manually or by use of the archive manager.