[RASMB] aggregates data analysis

[Borries Demeler]

Joris,

as others have pointed out already, a picture speaks a thousand
words here, and it would help to see some data. I think Walter is
right on target. If you don't get good fits under UltraScan, which 
model are you referring to? The least constrained model is the 
fixed molecular weight distribution. If you cover the range of species
well, it will *always* fit well and give you random residuals, provided
your data is a) at equilibrium, b) ideal c) there are no optical flaws
in the data (misfocussed interference are notorious here, since these
symptoms are easily missed). Even if you are using absorbance you still
can get optical artifacts as well.

-b.

> 
> Dear all,
> 
> I'm spinning a sample which shows -to me- surprising behavior. The raw data of a normal sedimentation equilibrium run of a protein looks always like a half-parabolic curve of the radius versus absorbance (x,y).
> However, this particular sample (protein) shows a different behavior in that the raw data look like a straight line taking off halfway the liquid column (absorbance almost zero) in the sector to the bottom of the cell (absorbance 0.7). Same behavior is reproducible at three different concentrations and three different rotational speeds (10,20,30 krpm).
> How should I interpret these data? Ultrascan does not really 'like' this data as input (although fitting is possible, the residuals look very bad, the fitted Mw -to a single species- is around 280kDa).
> Is this how a heterogeneous aggregating sample should look like?
> Thanks in advance.
> 
> Best wishes,
> 
> Joris Beld
> ETH Zürich
> Switzerland
> _______________________________________________
> RASMB mailing list
> RASMB at rasmb.bbri.org
> http://rasmb.bbri.org/mailman/listinfo/rasmb
>