Significant improvement of Oligomer fit?

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)
Post Reply
Posts: 1
Joined: 2017.09.27 13:17

Significant improvement of Oligomer fit?

#1 Post by linear » 2017.10.03 13:55


I work with two proteins that both have crystal structures available.
From SAXS data, protein A is monomeric in solution, while protein B is present as a trimer that is in a concentration-dependent equilbrium with higher oligomers. The proteins form a 1:1 complex.

I have SAXS data of protein A and B measured individually and in a 1:1 mixture.
From the mixture data, I have used Sasref for generating a model of the complex that fits my data well (chi^2 1.1).

My experiments are performed in two different buffers, in one buffer it is evident from molecular mass that complex formation takes place. In the other, there is only a slight increase in molecular mass with increasing concentration that could either correspond 1) to protein B forming higher oligomers, or 2) to the formation of protein complex.

I would like to use Oligomer for discriminating between these two scenarios.
First, I run Oligomer with the SAXS curves of protein A and protein B (at relevant concentrations) as input. I fix the ratio between the two proteins, i.e. Oligomer has 0 degrees of freedom.
Secondly, I run Oligomer with the Sasref model of the complex as input together with the two SAXS curves of protein A and B and a fixed ratio between them as input. Oligomer has one degree of freedom (volume fraction of the complex).

From the first to the second run, the chi^2 values improve suggesting that a complex is present in the mixture.
However, there is also an additional degree of freedom.

So my question comes down to:
How do I evaluate if an improvement in chi^2 is significant or not, also when taking the degrees of freedom into account?

Thanks in advance!

Post Reply