Equilibrium sedimentation data analysis software. Dmitry B. Veprintsev *, Nicholas W. Foster* and Alan R. Fersht, FRS *to whom correspondence should be addressed *dbv@mrc-lmb.cam.ac.uk *nwf@mrc-lmb.cam.ac.uk

Ultraspin is a new software package for the analysis of analytical ultracentrifuge equilibrium sedimentation data. The software can read .ra and .ip files directly, and has some limited editing capability (i.e., deleting ranges of datapoints or splitting datafiles into 3 for 6-sector cells). It is also possible to load multiple (up to 20) datafiles at once. The datasets can be fitted to a number of models, ranging from a single-species model to complex association equilibrium. The main advantage is the ability to fit multiple datasets simultaneously using the global fit procedure, optimizing both global and dataset specific parameters. This allows for combination of data at different speeds, concentrations or wavelengths to be mixed, significantly (statistically) improving results of the fit. The software also has the ability to model mixed association interactions of proteins.
The software will operate under Windows®95, or 98 and may be used under Windows NT®. It can not be used on Macintosh systems. Please read the software licence agreement before downloading Ultraspin,
 

Data Manager Window.

The data manager window comes up when the program is started. Here you load the data, save the fit, select a dataset to fit in the single dataset models and select the fitting models. You also have access to all different parts of the program from here. Results of curve fitting will be displayed in this window.

Loading data/Clearing traces

Use File/Open menu item to load files. Ultraspin can read Beckman absorbance and interference data files (.ra and .ip) directly. You can read the files in (up to about 20 simultaneously, but this can potentially be increased). It is also possible to select multiple files (with Cntr + Shift + left mouse button).
A list of all loaded files will appear in the drop down list in the left top corner.

You can clear all traces and data structures by Graph/Clear all menu.

Panning and Zooming, Axes control, data point information.

Panning.
Keep the right mouse pressed and move the mouse. This will "pan" the chart, and the point where you originally clicked will "stick" to the mouse pointer.

Zooming.
Click left mouse button on the upper left corner and keeping the mouse button down move to the bottom right corner of the selection.
Unzooming:
Click left mouse button on the bottom right corner and keeping the mouse button down move to the upper left corner. This will restore the original size of the chart.

Axes control.
If you click on (or near) the axes the axes control dialogue will pop up (it control the axe which was clicked).
You can specify the button and top values for the axe, and the increment of the tick marks on the axes. You can also select auto scale to scale the chart automatically to show all data points.

Data point information.
Click on the data point ? the dialogue box will pop up with information about x, y values, its number and data series it belongs to.

Serie
Select the dataset you want to fit from the drop down list in the left top corner of the main window. This works for the models that fit only one dataset (i.e., Spec1_MW_fixedshift, Spec1_MW_varshift, Spec1_V_fixedshift, Spec1_V_varshift).

Selectmultiple
SelectMultiple is for selecting multiple loaded datasets for fitting , specifying the region to use in the fit, Rbottom for each dataset, and initial values for local (dataset ?specific parameters). (See Select multiple window below)

Fit
Select the model you want to fit data to from the drop down list in the right top corner of the main window. (see "Fitting window" below)

The data can be fitted to different models. Each model has its own set of variable fit parameters, which should have initial values specified. As with all non-linear fitting algorithms, the initial guess should be quite good sometimes.

Fitting Models available.

The following models are currently available, please see the data fitting web page for a fuller explanation:

Models  OneSpecie_Mw_fixedshift  Spec1_Mw_varshift  Onespecie_V_fixedshift  Spec1_V_varshift  Spec1_Mw_varshift_Mult  Spec1_V_varshift_Mult  Dimerisation_Mw_K_varshift_Mult   Dimerisation_K_varshift_Mult  Dimerisation_K_varshift_Mult_Specrum  Monomer_Dimer_Tetramer_K1_K2_varshift_Mult_Spectrum  Protein_DNA_2A_plus_B_AB_K1_varshift_Mult_Spectrum  Protein_DNA_NA_plus_B_AB_K1_varshift_Mult_Spectrum  Nmerisation_K_varshift_Mult  Noninteracting_2_Mult  Noninteracting_3_Mult  Noninteracting_2_Mult_fixedshift  Noninteracting_3_Mult_fixedshift  nA_plus_B_K1_AnB_K2_AnB_m_fixedshift_Mult_Spectrum  nA_plus_B_K1lock_AnB_K2_AnB_m_fixedshift_Mult_Spectrum  A_and_AnB_K2_AnB_m_fixedshift_Mult_Spectrum

Terminology Onespecie - Single Species Spec1 - Single  Species Dimerisation - 2A -> A2 - Please note nmerisation provides more robust fitting Monomer_Dimer_Tetramer - 4A -> 2A2 -> A4 Protein_DNA - nA + B ->AnB Nmerisation - nA ->An Noninteracting - A1,A2,A3 K - Association constant Mw - Floating Molecular weight V - Partial specific constant Varshift - Floating baseline fixedshift - Fixed baseline Mult - Multi data fitting Spectrum - Multiwavelength fitting A_plus_B - nA + B -> AnB

Global and local parameters are shown on the main fitting window.  If fitting to more than a single dataset then local parameters for each dataset are entered in the dataset-selection window. (see below). The window also displays the fitting results

For the models which simultaneously fit multiple datasets (using GlobalFit), you have to select the required loaded datasets on the SelectMultiple window (under the button SelectMultiple).
Select the dataset in the left box, and then press -> button. To unselect dataset, select it in the right box, and click <- button. You can select few datasets simultaneously using combination shift-left mouse button to select all datasets between already selected and next to select, or Ctrl-left mouse button to select individual datasets.
 

.

menu: Data/AbsorbanceSpectrum

For all associating models (homo- or hetero-oligomerisation), it is neccesary to know the extinction coeeficient of the compound at a given wavelength in order to convert OD values into the molar concentration. There are EditBoxes on the main Fitting window for extinction coeeficients of A  (1) and B (2) called ExtCo and ExtCo2.

The ext. coefficient is calculated for every dataset as follows

Ext(wavelength)= ExtCo*AbsSpectrum(wavelength)*Path

where ExtCo (or ExtCo2) is from main fitting window, AbsSpectrum(wavelength) is from the 1st or 2nd column of the AbsSpectrumForm and path is specified under Settings/Constants (defaults to 1.2 cm).

GetWavelengths will show the wavelenghts which  are used in currently loaded datasets.
LoadSpectrum (into column 1 or 2, correspondingly) allows a 2 column ascii file (wavelength - value) to be read in - so you dont have to type it in every time.

Typically, I have normalised to 1 absorbance spectra in the file(s), and specify the ExtCo (ExtCo2) at the maximum on the main fitting form. If only one wavelength has been used, then putting 1 is all you need (as in the example shown).
 

© MRC (UK) Centre for Protein Engineering 2000