[RASMB] difference of p = 0.95, 0.68 and 0.55,

[Borries Demeler]

> Hi all
> 
> In Peter's absence let me make one point which he (very correctly) makes
> concerning estimation of s values and relative concentrations. Which is that
> the c(s) distribution for each individual species in a mixture is not a
> gaussian distribution. Hence one should not do 'peak fitting' by the usual
> algorithms, in ORIGIN or anything else. As Jack very correctly says, it is
> numerical integration over appropriate ranges which you need.
> 
> What is valid to a very good approximation is to fit multiple gaussians to
> the (least squares in SEDFIT) g(s) profile. Basically as per Walter
> Stafford's original approach.  In our own experience we find this gives a
> more objective description of such systems, albeit - since it is better at
> resolving closely related species including dimers of lower M (i.e. more
> rapidly diffusing) monomers - the outcome can be rather less flattering to
> the perceived 'quality' of one's precious preparation!

I can also suggest to look at this with the van Holde - Weischet analysis,
which gives an integral distribution of the concentrations that can be
directly read off the distribution plot. One advantage of the van Holde
- Weischet method is that the diffusion correction has already been
applied. This will result in a better resolution of multiple species than
g(s) distributions. Like Walter Stafford's method, this approach is also
independent of any user-bias resulting from preconceived s-value ranges
for the C(s) method or from a single f/f0 value for a large range of S-values.

In general it is a good idea to look at your data with multiple methods
and compare the results to make sure there aren't any conflicts in the 
interpretation.

Cheers, -Borries