Problem Solved - Short Summary
-------------------------------
Dear all,
First of all, a big thank you to everyone who devoted their precious time to help figure this out!
Good news, Refmac (tested with v.5.7.0032) does handle alternate sugar conformations correctly.
As Thomas Lütteke indicated, I have inadvertently tilted the sugar plane of conformation 'B' the wrong way (relative to
the Asn) causing an alpha-linkage instead of an beta-linkage. The conformation 'A' was correct (beta).
The monomer database denominates
beta-D-N-Acetyl glucosamine as NAG and alpha-D-N-Acetyl glucosamine as NDG.
However, Refmac dyamically assigns the alpha or beta linkage from the provided geometry rather than from the name.
So, the NAG-ASN LINKR record can describe either an alpha or a beta link, something which I did not expect.
I have now refined the A conformer and the B conformer separately at first, made sure that both geometries were beta and
then used the LINKR records and alternate conformation setup as described.
(Herman Schreuder pointed out that the linkage can also be described with standard PDB LINK records, instead.)
Everything works now, the sugars are no longer pushed apart.
(However, the B conformation seems to fit the electron density better than A, so there is only one conformation after
all...)
Cheers,
Markus
--
Markus Meier, Ph.D.
University of Manitoba
Department of Chemistry
144 Dysart Road
Winnipeg, MB, R3T 2N2, Canada
|