R Sigma Value

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
Message
Author
thd1
Member
Posts: 5
Joined: 2013.11.05 20:16

R Sigma Value

#1 Post by thd1 » 2015.07.27 17:13

Hey Everyone,

I ran EOM 2.0 on a protein that we know has a region of disorder and is pretty flexible, but I was curious about the high R sigma value I got. I know that R sigma should be greater than 1 for cases in which R flex is greater than that of the random distribution; however, I have never seen vales reported in the literature that are as high as the one I obtained in my analysis. The fit to the data seems to be pretty good and I got a chi square of 1.094. Is this high R sigma alright, or does this indicate the presence of artifacts?


Thanks,
Ty
Attachments
Screen Shot 2015-07-27 at 10.01.58 AM.png
Screen Shot 2015-07-27 at 10.01.58 AM.png (58.19 KiB) Viewed 3605 times
Screen Shot 2015-07-27 at 9.54.19 AM.png
Screen Shot 2015-07-27 at 9.54.19 AM.png (49.12 KiB) Viewed 3605 times
Screen Shot 2015-07-27 at 10.45.32 AM.png
Screen Shot 2015-07-27 at 10.45.32 AM.png (44.41 KiB) Viewed 3605 times

User avatar
Hayds
Active member
Posts: 101
Joined: 2008.05.21 19:01
Location: EMBL, Hamburg

Re: R Sigma Value

#2 Post by Hayds » 2015.08.05 05:17

Hi thd1,

I'm a bit interested in why the Rflex is higher for the selected ensemble than the pool. Does it make sense in this particular case? I assume you are using rigid bodies and a flexible region between?

Haydyn

thd1
Member
Posts: 5
Joined: 2013.11.05 20:16

Re: R Sigma Value

#3 Post by thd1 » 2015.08.10 14:22

Haydyn,

Thanks for the response. We believe this protein is highly flexible, so we thought it did seem reasonable to have Rflex higher for the ensemble than the pool. And we currently do not have any rigid bodies to use as inputs in the modeling.


Ty

emartin1
Active member
Posts: 43
Joined: 2012.08.06 21:22

Re: R Sigma Value

#4 Post by emartin1 » 2015.08.29 01:23

So, if I understand correctly, you're modeling the entire sequence as being flexible? Is this known to be the case? Admittedly, the data you show looks pretty featureless, but I'm curious as to how a Kratky plot looks. If you have well defined folded domains, this might not be the best approach for modeling the data. I can see how the right linear combination of unfolded conformers could accurately recreate the SAXS curve, but if the starting pool doesn't mimic reality you have a problem.

thd1
Member
Posts: 5
Joined: 2013.11.05 20:16

Re: R Sigma Value

#5 Post by thd1 » 2015.09.02 20:44

Thanks for the response. The dimensionless Kratky does indicate the protein is relatively non-globular (my protein is in blue compared to our lysozyme standard in green). We also have a CD spectrum that shows a mix of random coil and alpha-helical characteristic. Currently we think the center of the protein is a folded domain, while the majority of the N and C termini are disordered.
Attachments
kratky.pdf
(53.52 KiB) Downloaded 211 times

Post Reply