Dear Roberto,
If you are using shelxc/d/e, it is best to read the XDS.ASCII.HKL file
DIRECTLY into shelxc. This is what hkl2map does.
When you run shelxe (from the command line or hkl2map), you usually need
to try both hands of the heavy atom substructure. If the 'contrast' for
one hand is significantly better than for the other, then it is the
solution. Sometimes (e.g. because the heavy atom substructure is
centrosymmetric, e.g. one heavy atom in a polar space group) both
solutions have the same contrast and both are correct. However the
poly-Ala trace in shelxe will only work for the correct hand because
proteins are chiral. A particularly useful check is the CC against the
native data for the poly-Ala trace, if it is higher than 25% you have
solved the structure. This also applies when shelxe is used to expand
from a poor MR solution.
The phases from shelxe are already density modified, it is NOT a good
idea to density modify them again. You should simply read the .pdb and
then the .phs files from shelxe into Coot and look at the map yourself.
If it makes sense then you can either dock the side-chains yourself
using the facilities in Coot, or use Buccaneer to do this automatically.
shelxe does however currently require native data to relatively high
resolution (at least 3A and better 2A).
Best wishes, George
On 06/20/2014 05:54 PM, Roberto Valverde wrote:
Hi CCP4BB,
First time posting here, greetings!
I have two questions:
Question 1:
I am trying to use Shelx C/D/E pipeline to solve single anomalous
data. I processed the data in XDS, and scaled the data in aimless.
When I run shelx c/d/e pipeline I encounter this error:
#CCP4I TERMINATION STATUS 0 No ins file from shelxc
I assume this means there is no instruction file. How do define an
instruction file?
Question 2:
As an alternative I obtained phases from hkl2map, and wanted to
perform density modification using Parrot. I used the ha.pdb file I
obtained from hkl2map to define the atom sites. The program finished
and produced two possible solutions. Both maps look terrible, far
worse than the map hkl2map produced. What am I doing wrong?
Thank you,
Roberto
--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-33021 or -33068
Fax. +49-551-39-22582
