Why you should measure in Intensity mode and not in Absorbance mode
The importance of collecting velocity data in intensity mode:
When designing AUC experiments for UV optics it is highly recommended to measure in intensity rather than absorbance mode. Unlike absorbance data, intensity data is not convoluted with the random noise acquired in a reference measurement, producing better fits with lower RMSD values. In the process of collecting absorbance data in the Beckman XL-A, the reference channel data is subtracted from the sample channel data, thereby convoluting the random noise from the reference channel with the random noise from the sample channel. Whenever two noisy vectors are added or subtracted from each other, the noise level increases by the square root of 2, or a factor of ~ 1.4. The UltraScan software used in our laboratory is perfectly capable of accurately calculating the baseline levels of any experimental dataset using established algorithms, so a reference subtraction as done in absorbance experiments is no longer needed.
There are other advantages: In intensity mode both channels are scanned into separate datasets, making it possible to double the capacity of the instrument by using both channels for samples. There is only one catch: The OD in the reference channel must be less than 0.5 OD (as detected in a 1.2 cm pathlength cell), regardless of wavelength, to prevent a resetting of the photomultiplier gain set. If the total OD (including baseline absorbance compared to ddH2O) is less than 0.5 OD, the gainsets will not be changed. A gainset adjustment in the XLA during the experiment has important consequences: The data will be rescaled at a different scale, the baselines will shift, and the entire experimental data can no longer be globally analyzed in a single fit. Each dataset acquired at a different gainset needs to be analyzed separately (although the individual datasets can still be combined to be analyzed in a global fit). But this is tedious and inconvenient.
This limitation actually works in our favor: For many proteins it is interesting to study the protein at different concentrations. If a protein is reversibly self-associating, measuring a 0.3 and a 0.9 OD concentration under the same conditions will reveal mass action effects and tell us if the protein is self-associating. Hence, it is advisable to always measure at least two concentrations of each sample (0.3 and 0.9 OD, as measured in a standard 1.0 cm pathlength cuvette), with the 0.3 OD sample placed into the reference channel, and the 0.9 OD sample into the sample channel.