EOM - poor fit to data

Linear (OLIGOMER), and non-linear (MIXTURE) analysis, singular value decomposition (SVDPLOT), addition of missing fragments (BUNCH, CORAL), analysis of flexible systems (EOM/RANCH & GAJOE), flexible refinement of high-resolution models (SREFLEX)
Message
Author
Alex
Active member
Posts: 692
Joined: 2007.08.09 21:10
Location: Planet Earth

#16 Post by Alex » 2011.10.04 18:59

Jak wrote:Hi Alex,

Thanks for your help on this. I think background probably is the issue with Crysol.

Back to EOM - the reason we are trying to persevere with EOM is that we can influence conformer populations by changing sample conditions. But we should (and need) to relate the differences in populations under the new conditions back to the population(s) in the original conditions.
Here I saw only 1 condition. Do overall parameters change upon different conditions (rg, dmax, excluded volume)? Do curves look different at same concentrations under different conditions?
Jak wrote: Any ensemble, even if it is made up of one structure chosen several times, needs to fit the scattering curve if we are to accept it as plausible.

Can you think of any reasons why the fit doesn't work in the 0.1-0.2 region? Is EOM trying to insist on a domain feature that doesn't actually exist in the real curve?
There is no issue with CRYSOL here. Maybe buffer scattering between 5 and 10 mg samples is very different, hence resulting subtracted data looks different from one at lower concentrations. So, check your buffer first.

Jak
Active member
Posts: 25
Joined: 2007.10.01 13:01
Location: UK

#17 Post by Jak » 2011.11.07 15:21

Here I saw only 1 condition. Do overall parameters change upon different conditions (rg, dmax, excluded volume)? Do curves look different at same concentrations under different conditions?
Sorry, I only just saw this reply.
Yes the curves are consistently different.
Jak wrote:

Any ensemble, even if it is made up of one structure chosen several times, needs to fit the scattering curve if we are to accept it as plausible.

Can you think of any reasons why the fit doesn't work in the 0.1-0.2 region? Is EOM trying to insist on a domain feature that doesn't actually exist in the real curve?

There is no issue with CRYSOL here. Maybe buffer scattering between 5 and 10 mg samples is very different, hence resulting subtracted data looks different from one at lower concentrations. So, check your buffer first.
I was still referring to the EOM fit. As you say, the Crysol problem was sorted. The original issue is that EOM should be able to fit the data but it puts in that extra feature between 0.1-0.2 or so, and doesn't really fit after that either.

Post Reply