Dear all,
Thank you for the response. I will try to explain it more precisely.
The molecule of interest is a duplex with 9 nt length, but is paired on
the length of eight bases, with overhangs at ends. Molecules form a
parallel strings across the crystal lattice, parallel to C-axis, because
of these stacked overhangs. The structure was solved by MR using Molrep.
Trials using Phaser were failed. The initial model was obtained by
ZN-SAD. Refinement was dome for space group P43212, with cell parameters
31.96 31.96 95.07 90 90 90, with one duplex molecule per AU.
Schreuder, Herman /DE wrote:
At this resolution, one sees many amino-acid side chains with
alternative conformation, so it might be a good idea to test if this is
also true for nucleotides.
Dear Herman,
I'm working on some protein ultra-high resolution structures (around 1.0
A or higher), and alternative conformations are nicely visible on
electron density maps. In this case, there is visible almost all
molecule, when you switch contouring to 2 sigma or lower on Fo-Fc maps,
so I think, in this case it's not the same situation.
Matthew Snee wrote:
Maybe test the spacegroup with Zanuda, and reprocess with the most
likely lower symmetry group.
I guess the stats should improve if you identify a pseudo symmetry
operator that is currently being treated as a true symmetry operator?
Eleanor Dodson wrote:
Sometimes ghost like this mean there is a spacegroup error - absences
can be the result of the non-crystallographic translation and not be
truly indicitive of the spacegroup. What is the possible spacegroup and
what is the NC translation vector?
Dear Matthew and Eleanor,
I run Zanuda on my datasets, and the output (which is below) suggested,
that spacegroup is right chosen.
Step 1.
R-factors for the starting model.
Transformation into a supergroup.
---------------------------------------------------------------------
| Subgroup | Spacegroup | R.m.s.d. | Refinement in tested group |
| | | from the |--------------------------------|
| Ref | | starting | Rigid | Restrained |
| | | model, A |----------|---------------------|
| | | | R | R | R-free |
|----------|------------|----------|----------|----------|----------|
| >> 10 | P 43 21 2 | 0.0002 | -- | 0.5107 | 0.4871 |
| 10 | P 43 21 2 | 0.0002 | -- | -- | -- |
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Step 2.
Refinements in subgroups.
There are 8 subgroups to test.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| >> 10 | P 43 21 2 | 0.0002 | -- | 0.5107 | 0.4871 |
---------------------------------------------------------------------
| 1 | P 1 | 0.0883 | 0.5252 | 0.4985 | 0.4883 |
| 2 | C 1 2 1 | 0.0828 | 0.5447 | 0.5006 | 0.4877 |
| 3 | P 1 21 1 | 0.0824 | 0.5367 | 0.5018 | 0.4921 |
| 4 | P 1 21 1 | 0.0789 | 0.5292 | 0.4971 | 0.4846 |
| 6 | P 21 21 21 | 0.0956 | 0.5380 | 0.5064 | 0.4929 |
| 7 | P 43 | 0.0935 | 0.5183 | 0.4952 | 0.4835 |
| 9 | C 2 2 21 | 0.0908 | 0.5435 | 0.5042 | 0.4910 |
| 10 | P 43 21 2 | 0.0855 | 0.5427 | 0.5097 | 0.4913 |
---------------------------------------------------------------------
| << 7 | P 43 | 0.0935 | 0.5183 | 0.4952 | 0.4835 |
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Step 3.
Refinement of the best model.
Candidate symmetry elements are added one by one.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| >> 7 | P 43 | 0.0935 | 0.5183 | 0.4952 | 0.4835 |
---------------------------------------------------------------------
| 1 | P 1 | 0.0927 | 0.5293 | 0.4991 | 0.4950 |
| 8 | C 1 2 1 | 0.0848 | -- | 0.5017 | 0.4906 |
| 9 | C 2 2 21 | 0.0871 | -- | 0.5059 | 0.4928 |
| 10 | P 43 21 2 | 0.0919 | -- | 0.5180 | 0.5109 |
---------------------------------------------------------------------
| << 10 | P 43 21 2 | 0.0919 | -- | 0.5180 | 0.5109 |
---------------------------------------------------------------------
R-factor in the original subgroup is (almost) the best.
The original spacegroup assignment seems to be correct.
According the non-crystallography translation vector, there is an output
from xtriage:
-----------------------------------------------------
| XYZ | height | p-value(height) |
-----------------------------------------------------
| 0.000, 0.000, 0.334 | 53.049 | 4.456e-05 |
| 0.000, 0.000, 0.167 | 28.966 | 1.681e-03 |
| 0.000, 0.000, 0.500 | 27.692 | 2.102e-03 |
-----------------------------------------------------
Jon Cooper wrote:
Hello, have you tried anisotropic B-factor refinement? It is usually
very good at cleaning-up the difference map. At that sort of
resolution, you may still have some way to go in the refinement since
the R and R-free usually go to about half of what you have, at least
with proteins.
Dear Jon. Yes, I used anisotropic refinement as usually for this sort of
resolution (1.4 A or better). But it looks too high for this small
object and very high resolution, in my opinion.
Best regards,
Rafal
########################################################################
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/