Up ] experiments ] parameters ] partial specific volumes ] [ tutorial for getting started ] configurations ] sedimentation equilibrium ] sedimentation velocity for reactive systems ] sedimentation isotherm analysis ] isothermal titration calorimetry ] surface binding isotherms ]

Typical workflow for using SEDPHAT

These are the typical steps for a Sedphat analysis:

0) sometimes preparation of files:  convert DLS raw data into dlsdat files with Options->Loading Options->Import DLS Data" of SedfiT, or rename data with Options->Loading Options->Save Raw Data" of SedfiT to make them easier to identify. 

1) generate an xp-file for each experiment.  Here, you load the raw data, using the functions "Data->Load New ... Data".  An experiment box will appear prompting you to enter the experimental conditions, like buffer density, viscosity, partial-specific volume of the molecules, noise and baseline parameters, extinction coefficients (see also Experimental Parameters).  Enter meniscus, bottom, and fitting limits now (usually graphically, like in SedfiT, but the meniscus and bottom can also be entered numerically in the experiment box).

2) save this xp file, using the function Data->Save Experiment. This will allow you later to reload exactly the same experiment, with exactly the same selection of raw data, and the same settings (you'll not need to enter any of the previous values again).  This will generate an xp-file on your disk.  Note that this will not save the raw data, but only a text-file with the pointer to the data, and the additional information.  For details look at Experiments.

Note:  It is recommended to save the data are in a path that has 'white-space' characters:  'c:\My Data' is bad, but 'c:\MyData' or 'c:\My_Data' is OK (likewise, 'c:\My Documents\...' or 'c:\XLA Data' are bad).  Also, don't use the Desktop!  SEDPHAT should be able to deal with white-space characters from version 4.1b and later. 

repeat steps 1 - 2 for all your data channels you want to analyze.

3) load several (or sometimes only one) xp-file, using Data->Load Experiment .  Note you can also drag and drop xp-files into the Sedphat window.

4) choose the analysis model.  Initialize, according to the model, the local and global parameters, establish links to combine local parameters into shared parameters (maybe meniscus, bottom, local concentration, extinction coefficients), set appropriate constraints in the model, such as mass conservation, molar ratios of loading concentration.

5) save the Sedphat configuration with Data->Save Configuration. This will allow you to reproduce later the exact combination of data and model.  If you are planning to do several alternative analyses involving the same data, then answer yes to the option to include new xp-files.  See more details on this in Configurations.

6) execute the model using the Run function.  This is a way to evaluate the quality of your initial guesses of the non-linear parameters. 

maybe go back and repeat steps 4 - 6 to refine your starting guesses until the model functions are at least in the rough vicinity of the data

7) optimize the parameters with the Fit function.  (For some types of fits, I prefer Simplex method, for others the Marquardt-Levenberg.  You may want to do both sequentially to make sure you have really found the best fit (Options->Fitting Options)

8) save the results by saving (or updating Data->Update Configuration [control - U]) your configuration and save all experiments (Data->Save All Experiments [control - A]).  Note that saving the experiments at this point also saves the local parameters.  (The latter may conflict in different alternative analyses, which is why each configuration should have their own copy of the xp-files.)  This step allows you to precisely recreate this analysis later from scratch by double-clicking the sedphat-file in your file-manager.

mabye go back to repeat steps 4 - 8 to produce alternative analyses, relax or increase constraints, make controls (like 'would the data fit assuming there was no binding...', etc., )

9) document the analysis: make a screenshot (Alt-PrntScrn), paste it into powerpoint, write comments to it.  Note that you may want to copy the Sedphat screen also in the state where it shows you the numerical details of all parameters, you can recall these by Display->Show Last Fit Info Again.  Note also that the title bar of the Sedphat window contains the path of the current configuration file which you can use to recreate the analysis.

10) export the results and copy the data tables to a spreadsheet in your favorite plotting software.


If you have problems, have a look at the troubleshooting and frequent questions page.