[RASMB] Photomultiplier tube disturbances?

[Borries Demeler]

Troy,

Do you notice any sudden drops/changes in the intensity over the course of
the experiment? When you pick the reference and plot the intensity plot over
the course of the experiment, is that intensity vs. scan number plot 
relatively constant, or are there jumps in the intensity of more than 5% from
scan to scan? It would be helpful if you could send that plot along.

The only other thing I can think of is that one of your references that contains
sample has an absorbance that together with the DNA absorbance puts you
over the allowed maximum OD in the reference cell. In that case, run all
samples against water and see if the problem disappears. Having too high 
OD in one of the reference cells will change the gain setting on the PM tube
and give the kinds of changes in data scaling we seem to have here.

While that problem is easily correctable by using a lower OD or water in
the reference, the former is more difficult to eliminate. If you have
non-constant (relative) intensity with big jumps in the intensity over time,
you need to call service and have the machine repaired. The intensity
should not change much over the course of the experiment.

There are two ways to fit this data:
The best approach would be to separate the scans into separate runs and
fit them separately, since they may be scaled differently. If this is 
just a constant offset (it doesn't appear that way from your magnified 
scan image), then you could also try to fit the different as time-invariant noise.

-Borries

> 
> Hello,
> 
> I am performing sedimentation velocity experiments in a XL-I in
> intensity mode, scanning at 260 nm. My samples are a protein-DNA
> nucleosomal array complex.  I am running a four hole rotor with cell 1
> sector 1 as my ddH20 blank, the remaining 5 sectors contain a DNA
> concentration equilavent to ~0.2 absorbance units.  I am spinning the
> rotor at 20K rpms for about 2.5 hours.  The intensity of the machine
> at 259 nm is relatively high at ~45000 units.
> 
> My issue seems to be with the photomultiplier displacing one or more
> scans (typically a decrease in magnitude) during the course of the
> run.  I do not believe this is from sample sedimenting as the decrease
> in magnitude is the same and at the same scan number for both sectors
> in a cell, but at different scan numbers for different cells.  The
> downward shift doesn't always happen for each cell either.  It may
> occur in just 1, 2, or all 3 cells.  I can have a run where it only
> occurs in one cell and the other two cells are fine.
> 
> The images attached are from the Ultrascan software that converts the
> intensity data to pseudo-absorbance, and a zoomed in visual of the
> psuedo-absorbance data. I would be more than happy to provide more
> information if my inquiry is to vague!
> 
> If this is due to the high intensity at 259 nm, would using scanning
> at a wavelength like 262 nm (~22000 units) circumvent this problem?
> Any other suggests, comments, concerns would be greatly appreciated.
> 
> Thanks for your time,
> Troy Sorensen
> 
> Hansen Lab
> Department of Biochemistry and Molecular Biology
> Colorado State University, Fort Collins CO