Hi all
I'm working on a structure of an RNA. Phases were obtained via SAD and
the structure was refined to 21/26 % on R and R-free with a resolution
of 2.8A, single molecule in the ASU. 7 residues were missing in a
disordered loop out of a total of 54 residues in the asu. The cell was
R32, dimensions 105.3598 105.3598 107.7804 90.0000 90.0000
120.0000. Xtriage showed decent statistics for this dataset:
Wilson ratio and moments
Acentric reflections
<I^2>/<I>^2 :2.060 (untwinned: 2.000; perfect twin 1.500)
<F>^2/<F^2> :0.779 (untwinned: 0.785; perfect twin 0.885)
<|E^2 - 1|> :0.759 (untwinned: 0.736; perfect twin 0.541)
Centric reflections
<I^2>/<I>^2 :3.131 (untwinned: 3.000; perfect twin 2.000)
<F>^2/<F^2> :0.653 (untwinned: 0.637; perfect twin 0.785)
<|E^2 - 1|> :0.937 (untwinned: 0.968; perfect twin 0.736)
We were able to obtain crystals that shot to about 2.0A, again in R32,
with the unit cell dimensions 103.5 103.5 108.888 90 90 120.
Refinement of the low resolution structure into this high resolution
dataset proves to be problematic. The last 7 residues cannot be built
without building into a symmetry related molecule yet there is
uninterpretable positive density in the 2Fo-Fc and Fo-Fc maps for
these last residues. The refinement statistics stall at at 26.8 and
30.6 (R and Rfree using the same test set as the low res case except
extended for the higher resolution). This seems rather high
considering that only 7 residues are missing. Xtriage shows an
elevated wilson statistic for this dataset:
Wilson ratio and moments
Acentric reflections
<I^2>/<I>^2 :2.168 (untwinned: 2.000; perfect twin 1.500)
<F>^2/<F^2> :0.772 (untwinned: 0.785; perfect twin 0.885)
<|E^2 - 1|> :0.762 (untwinned: 0.736; perfect twin 0.541)
Centric reflections
<I^2>/<I>^2 :3.376 (untwinned: 3.000; perfect twin 2.000)
<F>^2/<F^2> :0.616 (untwinned: 0.637; perfect twin 0.785)
<|E^2 - 1|> :1.059 (untwinned: 0.968; perfect twin 0.736)
I'm particularly concerned that <I^2>/<I>^2 for both centric and
acentric reflections are elevated above the untwinned case, suggesting
the symmetry/space group is incorrect.
R32 has C2 and R3 underneath it but reintegrating into those datasets
and MR of the best model doesn't change the refinement statistics or
the wilson ratio statistics, (stuck in high R/Rfree and greater than 2
and 3 for the acentric and centric reflections).This is refining with
NCS or appropriate twin laws for those space groups.
I would love to entertain comments.
Thanks
FR
--------------------------------------------
Francis Reyes M.Sc.
215 UCB
University of Colorado at Boulder
gpg --keyserver pgp.mit.edu --recv-keys 67BA8D5D
8AE2 F2F4 90F7 9640 28BC 686F 78FD 6669 67BA 8D5D
|