Data extrapolation required?

Interactive and automated data processing tools (PRIMUS, GNOM, AUTORG).
Scattering from simple bodies (BODIES), peak analysis (PEAK), data plotting (SASPLOT) etc.
Post Reply
Message
Author
dapereir
Guest
Posts: 1
Joined: 2018.06.14 17:41

Data extrapolation required?

#1 Post by dapereir » 2018.06.14 18:13

Hello,

I am comparing the same protein (2 domains+linker) with point mutations. I measure at batch mode using >6 concentrations collected from SEC performed previously, and protein behaves well (I mean no obvious aggregation and is surely monomeric). I want to verify whether these mutations extend the protein, and whether it affects flexibility (visually checked via Kratky for now, not EOM yet).

QUESTION 1: the wild-type and mutant A are fine, different concentrations do not change Rg or Dmax, MW estimation is also within the error. However if I extrapolate this curve using mutant A, I see changes in Dmax (from average 10.7-11.0 for 6 concentrations down to 8 nm for the extrapolated one, which is quite substantial difference). I am not sure why that happens with the extrapolated curve, since I do not see clear signs of aggregation in the samples used for extrapolation, and buffer subtraction was properly done.

QUESTION 2: Regard mutant B, I see a linear correlation of Rg and Dmax depending on the concentration (higher concentration slightly superestimate these parameters). Thus in this case it seems there is a structural factor affecting the data. This can be minimised from extrapolation using the 7 curves from different concentrations or this data cannot be used? Or any other suggestion how to proceed please?

Since the extrapolated data for Mutant A provides different Dmax then the subtracted curves, I am not sure if I should compare only extrapolate data between the mutants, or if it's fine to compare a subtracted curve for mutant A vs. extrapolated for mutant B.

Below you can find the files for mutant A and B in case you can to have a look.
Thanks in advance.
Attachments
Mutant_B.zip
(140.63 KiB) Downloaded 11 times
Mutant_A.zip
(140.5 KiB) Downloaded 12 times

User avatar
AL
Administrator
Posts: 651
Joined: 2007.08.03 18:55
Location: EMBL Hamburg, Germany
Contact:

Re: Data extrapolation required?

#2 Post by AL » 2018.08.28 11:44

dapereir wrote:
2018.06.14 18:13
QUESTION 1: the wild-type and mutant A are fine, different concentrations do not change Rg or Dmax...
Mutant_A/S?.dat: from the rather narrow concentration range of 0.6-2.1 mg/ml I'd say there are concentration effects: the Guinier Rg increases with concentration from 2.6 to 3.0 nm. The extrapolated curve looks okay - apart from the very low angles (s<0.2 nm-1); if you remove the first 40 points you get a p(r) with Dmax=9.3 nm which fits the data well.
dapereir wrote:
2018.06.14 18:13
Since the extrapolated data for Mutant A provides different Dmax then the subtracted curves, I am not sure if I should compare only extrapolate data between the mutants, or if it's fine to compare a subtracted curve for mutant A vs. extrapolated for mutant B.
Start by comparing the background-subtracted data of similar concentrations: if you plot 0.6 mg/ml curves S8 and K8 you will see only a small difference at s < 0.26 nm-1; Guinier RgS8=2.6 nm, RgK8=2.7 nm - that's your difference, i.e. Mutant_A is slightly more compact than Mutant_B.
If you compare 1.3 mg/ml data S5 and K3 or 2.1 mg/ml data S1 and K1 you will see a similar difference at s < 0.4 nm-1 but the curves look same at higher angles - so nothing to add here apart from Mutant_B tends to aggregate with concentration slightly more than Mutant_A. Even your extrapolated curves look strikingly similar.

Post Reply