Hi Catherine, I think you meant to write "giving high R-factors".
To add to what Esko and Eleanor wrote: There is nothing wrong with your data or spacegroup or refinement; it is just that the tNCS produces two sets of reflections: half or your reflections are statistically "normal" (as if no tNCS were there), and the other half is quite weak (their intensities would be zero if the translation were 0,-0.5,0.5 which is not far from 0.072,-0.498,0.5). As a result, one half of your reflections has a relatively good agreement of Fcalc and Fobs; the other half agrees relatively worse because a) they have less weight in refinement, b) their measurement error is relatively higher than that of the strong reflections. The R-factor calculation knows nothing about this, it just uses all reflections and so you get something like an average R-factor which is of course higher than that of the strong half. I'd say just document this in the paper you write about the structure. It would be useful/interesting for such cases to have a program that calculates R-values separately for the two classes of reflections (it should be possible to do this with a script that employs CCP4 sftools, I think). HTH, Kay On Wed, 29 May 2024 13:48:46 +0000, Catherine Back <catherine.b...@bristol.ac.uk> wrote: >Hi, > >I have collected data for a protein which solved easily to 1.6 � using MR. >However, during data analysis and refinement it clearly has translational NCS. >According to Xtriage on Phenix: >Frac. coord.: 0.072, -0.498, 0.5 >Distance to origin: 68.798 >Height relative to origin: 51.479% >p-value (height): 5.577e-05 > >Wilson value and L-test suggests no twinning. > >It solved using Molrep in ccp4i2. P2 21 21. All stats look good, but even >after refinement (not fully and completely finalised yet) R= 0.24 and >Rfree=0.27. Also, it appears the 2 molecules in the AU have swapped a helix >with their respective symmetry partners. I'm pretty sure I now have the >correct space group, and tried various others with much less success. > > > 1. >Will these Rfactors be 'good'/low enough to publish for a dataset at 1.6�? > 2. >Should I be doing anything else during structure solving to alleviate the >effects of tNCS? (I did try using Phaser to solve the structure, but the stats >were much worse and even though 'tNCS is present correction factors were not >applied'. I'm not sure why. > >Many thanks, >Cat > > >Dr Catherine R. Back (she/her) > >Senior Post-doctoral Research Associate > >School of Biochemistry > >University of Bristol > >UK > > > > >######################################################################## > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing >list hosted by www.jiscmail.ac.uk, terms & conditions are available at >https://www.jiscmail.ac.uk/policyandsecurity/ > ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
