On 31/03/15 09:54, Jose Artur Brito wrote:
Dear All,
I was experiencing the same problem when reading pdb files with
duplicate numbers so I tried this fix that Paul suggested. However,
when I had that code line to the .coot file, here's what comes up:
Loading ~/.coot...(set-find-hydrogen-torsion 1)
Backtrace:
In /home/jbrito/.coot:
29: 0* (allow-duplicate-sequence-numbers)
/home/jbrito/.coot:29:1: In expression (allow-duplicate-sequence-numbers):
/home/jbrito/.coot:29:1: Unbound variable:
allow-duplicate-sequence-numbers
Gtk-Message: Failed to load module "canberra-gtk-module"
Gtk-Message: Failed to load module "canberra-gtk-module"
Gtk-Message: Failed to load module "overlay-scrollbar"
Gtk-Message: Failed to load module "unity-gtk-module"
Gtk-Message: Failed to load module "canberra-gtk-module"
I am running Coot 0.8.1 (Turtle Bay), under Ubuntu 14.04 LTS.
Any ideas what might be the issue here?
Cheers,
Jose
On 03/27/2015 07:13 PM, Paul Emsley wrote:
On 25/03/15 22:27, nih\mayerm wrote:
Deal Paul.
Recently I had an unusual case of dual occupancy of a ligand binding
site by two different ligands. I refined this by giving same resids
to the two ligands (here C1 in coordinates below), refining their
occupancy (in Phenix) - worked very well. Older versions of COOT
opened this PBD file just fine.
Now, when I open in coot 0.8.1 I get the following error, and file
cannot be opened.
ERROR 42 READ: Duplicate sequence number and insertion code.
LINE #12921
HETATM 8309 N BGLY C 1 9.972 61.351 43.716 0.24
9.17 N
I realize my solution to the dual occupancy issue is a bit of a
cludge (of which I was quite proud!) but I'm now forced to use an
older version of coot when working with these files. Is the change a
feature or a bug?
It was a feature that stopped a bug (crash in a ccp4 library IIRC) in
certain atom selections on chains that include duplicate sequence
numbers.
It seems that such files happen more often the I had realised and
have since made it a user-setable option:
Add this to your ~/.coot file
(allow-duplicate-sequence-numbers)
It's in the current 0.8.2-pre and and will be in 0.8.2 and later
releases. It's dangerous though.
Paul.
This feature is in 0.8.2-pre, revision 5611 and greater.
Paul.
