*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear Jeremiah,
Even if you could do it, I don't think you could justify that approach.
You're justified in using restraints to keep something where you know that
it should be (but don't have enough data to trust it not to float away),
but you're not justified in using restraints to keep something in the wrong
place. If you force the ligand to occupy the same position in the higher
resolution (but presumably radiation-damaged) structure as in the lower
resolution structure, the rest of the structure will be degraded in
accuracy, as compensating errors will be introduced.
One sensible aproach would be to refine against the higher resolution data
first, without restraining the ligand. This should give you the best model
for the vast majority of the atoms, and will give you an excellent starting
point for refinement against the lower resolution data. In the second
refinement, you should only have to adjust the model in a few places where
there was radiation damage during collection of the higher resolution data.
In your analysis, you would use the lower resolution structure to talk
about ligand interactions, and the higher resolution structure to talk
about the rest of the protein.
It ought to be possible to restrain a lower resolution structure to look
like a higher resolution structure, but off the top of my head I don't know
how to do this, or which program might allow this. This would be justified,
as long as you only restrain the parts that should be the same. This means
keeping an eye on gradient maps or difference maps for evidence that the
data demand that the lower resolution structure should differ in some
region, then relaxing the restraints in that region.
Randy Read
On Nov 17 2006, JEREMIAH R WAGNER wrote:
Hello All, I collected two data sets on a protein crystal. One was
collected using xrays from a rotating anode detector and the other with
synchrotron radiation. Both data sets are essentially the same, but the
synchrotron data set was collected at a higher resolution. However, I
have a covalently attached ligand that appears to shift its position in
the synchrotron data set presumably because of radiation breakage of the
bond between the protein and the ligand. This damage to the ligand bond
is not present in the rotating anode data set, and because of this we
would like to try to fix the position of the ligand/bond of the rotating
anode data set into the synchrotron data set and refine this "hybrid"
structure in refmac. Is there a way to prevent refmac from moving the
positions of a few certain atoms while refining the remainder of the
structure? Thanks, Jeremiah Wagner
