Refining a common occupancy (using one free variable) for a loop that is
in poor density is well worth trying. The loop may well have multiple
conformations and this may make it easier to find a second conformation in
the difference map. I also recommend refining the occupancies of any
selenium atoms present (e.g. by changing '11' to '1'), this allows for
partial incorporation and radiation damage.
Hydrogen atoms should be included in all final refinements, they cost
no extra parameters and can reduce the free R by 0.5 to 1.0%; the
antibumping restraints involving them (BUMP) are also useful. However I
do not include OH hydrogens because the stupid program sometimes puts
them in the wrong place (e.g. two on the same H-bond) and the combination
of the riding model and antibumping restraints can tear the structure
apart. For amide sidechains and histidines I would recommend checking the
conformations with the molprobity server before adding the hydrogens in
SHELXL. It is however less work to model all alternative conformations
before adding hydrogens with HFIX, the program will then set up the
disorder correctly for the hydrogens too. In such cases the disorder
should be modeled one atom further back than you can see in the maps; if
CG has two positions then there should be two pairs of alternative H-atoms
on CB, which the program will set automatically if you have included PART
1 and PART 2 alternatives for CB. This is appreciably more work to set up
later by hand if you add the hydrogens before modeling the disorder!
It is worth trying to make the waters anisotropic (with an ISOR restraint)
to see if this reduces R free significantly. If the .lst file gives NPD
warnings the restraints on the anisotropic atoms are too soft.
And I am surprised that after all your emails on the subject you still
can't spell 'SHELX'!
George
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-3021 or -3068
Fax. +49-551-39-2582
On Wed, 21 Mar 2007, U Sam wrote:
> I am looking for some advice.
>
> (1) In shelex what should I mention to refine occupancy.
> I have two molecule in asym unit.
> In A molecule residues 89-92 is present, but in B these residues are missing.
> So I believe in B these residues should not be with zero occupancy, although I
> donot find any prominent density (Fo-Fc). Occupancy could be anywhere between
> 0.0 to 1.0. How can I refine this parameter. Or, I should neglect this missing
> part of residues in B indicating a occupancy of 0.0 or keep a gap of these
> residues with no information including coordinates.
>
> Right now R1=14% and R1(free) =18%, without making water anisotropic.
>
> (2) I am using 1.4 A data. Should I refine water anisotropically ? If answer
> is yes, when.
>
> (3) Should I add hydrogen at this resolution. If yes, when should I do.
>
> Thanks
> Sam
>
> _________________________________________________________________
> The average US Credit Score is 675. The cost to see yours: $0 by Experian.
> http://www.freecreditreport.com/pm/default.aspx?sc=660600&bcd=EMAILFOOTERAVERAGE
>
|