Sorry for the delay, it's hard for me to do non-trivial mails while I'm
travelling.
On 23/11/2017 15:52, Toon Van Thillo wrote:
For my master thesis I am busy with neutron crystallography. For refinement, the software used is
/phenix.refine /and COOT for manual corrections.
However, it seems that deuterium atoms will not correctly refine because there are no restraints available
in the COOT dictionary.
Yes. It's using the standard Refmac dictionary.
Thus we set out to introduce restraints to COOT by creating cif files for each amino acid with restraints
for deuteriums.
This seems to work for perdeuterated proteins, but not for those that are
H/D-exchanged.
OK, makes sense.
The cause of this problem seems to be in the way H/D positions are defined: /phenix.refine/ assigns H and D
to alternative conformations A and B,
while the other part of the residue remains normal (eg. has no alternative
conformation assigned).
ATOM 1080 N ALA A 67 16.629 -3.415 9.989 1.00 7.92
N
ATOM 1081 CA ALA A 67 15.560 -3.601 9.012 1.00 10.82
C
ATOM 1082 C ALA A 67 14.894 -4.957 9.211 1.00 8.09
C
ATOM 1083 O ALA A 67 14.918 -5.812 8.329 1.00 6.06
O
ATOM 1084 CB ALA A 67 16.119 -3.459 7.599 1.00 7.64
C
ATOM 1085 HA ALA A 67 14.801 -2.842 9.167 1.00 9.05
H
ATOM 1086 HB1 ALA A 67 15.405 -3.583 6.954 1.00 9.12
H
ATOM 1087 HB2 ALA A 67 16.502 -2.575 7.490 1.00 9.97
H
ATOM 1088 HB3 ALA A 67 16.806 -4.129 7.454 1.00 9.88
H
ATOM 1089 H AALA A 67 17.253 -4.214 10.089 0.50 9.73
H
ATOM 1090 D BALA A 67 17.253 -4.214 10.089 0.50 9.73
D
Yes.
The restraints we introduced in the cif files do work when deuterium and the corresponding hydrogen have not
been assigned to an alternative conformation.
Also makes sense.
When they are, they (both hydrogen and deuterium) refuse to refine. Does
someone know a way around this problem?
Phenix neutron cif files are not well handled by coot. The expectation that there is only one dictionary
atom for any given atom name is deeply entrenched in the way Coot deals with cif files, and stores and uses
dictionary information. If it had been straightforward to fix this I would have done (Pavel had told me
about this a while ago) - but it's not :-(. I did make Deuterium atoms pink though.
Paul.
