Orient uses a Monte Carlo approach to fit a molecular fragment on a site. Typically, several dozen or hundred random orientations are tested. The orientation that gives the best R1 is saved.
Download the correct version for your operating system:
Place the executable in the system path. The directory containing your ShelXL executable might be convenient.
Optionally, a test structure is available here: cluster.ins
cluster.hkl. Place these files anywhere and call Orient from
your command line: $ orient cluster. You should see 200 ShelXL trials. A correct orientation will give R1 ~3.6%.
The executables were compiled on Windows 10 Home, MacOS 10.14.6, and Ubuntu 16.04 respectively. Please contact the author with any problems running them or for other platforms to add.
$ git clone https://github.com/dwpaley/orient
$ chmod +x orient/orient.py
$ ln -s $PWD/orient/orient.py /usr/local/bin/orient
Orient is called with
$ orient <name>
where the working directory contains <name>.ins and <name>.hkl.
The ins file is a standard ShelXL format file but should contain the following:
-
A FRAG fragment in orthogonal coordinates (as obtained from the Idealized Molecular Geometry Library, http://xray.chem.wisc.edu/Projects/IdealizedMolecularGeometry.html, or generated by Olex2).
-
A block of atoms to which the orient fragment will be fitted. You are responsible for setting the part and occupancy of these atoms and any other settings, such as RESI name, that you wish to include. The occupancy must be set on a PART line. The coordinates may be filled with zeroes.
-
Surrounding these atoms, add instructions for the Orient run. Start with:
!!orient x y z [--shift=0] [--trials=100] [--frag=17] [--uiso=.05] [--afix=6]
and end with:
!!oend
The arguments x, y, and z give the centroid of the orient fragment. If they are
omitted, the current centroid of the atoms between !!orient and !!oend is
used.
The optional arguments are the following: shift is a random translational shift of n angstroms (default 0) in case the centroid is not known accurately. Trials is the number of random orientations to test. Frag is the number of the FRAG fragment that will be oriented. (This may be changed in case you are using other FRAG fragments at the same time.) Uiso is the isotropic adp for each atom in the search fragment and may be set to a free variable, i.e. -u21, to set a group isotropic ADP. Afix is the /n/ component of the AFIX code that will be used for the search fragment. The default, 6, gives a rigid-body refinement, but 0 may be useful for a flexible fragment (maybe with restraints).
The number of trials needed depends on the size of the search fragment, the point symmetry of the fragment, and the point symmetry of its crystallographic site. In a very difficult scenario, with a molecule in point group C1 on a general position and assuming a molecular "radius" of 4 A, it may require about 2000 trials to reach a 90% chance of a successful orientation. This number may be scaled directly by the symmetry of the molecule, the symmetry of the site, and the third power of the molecular radius. Thus, for toluene (point group C2v, "radius" ~2 A) on an inversion center, we might expect 90% success in about 30 trials. However, the convergence of the orientation depends on the details of the molecule and site.
Set a cgls refinement with a small number of cycles; 10 typically works well. Besides the centroid coordinates, it is often fine to omit the other run parameters and use the defaults.
The following lines, when added to an ins file and called with $ orient <name>,
will run 30 trials to orient a toluene molecule on the inversion center
xyz = 0.5 0.5 0.5.
FRAG 17
C1 1 1.198402 -0.033739 3.629224
C2 1 1.201835 -0.000101 2.122450
C3 1 0.001012 -0.000301 1.398484
C4 1 0.000813 -0.000261 0.002140
C5 1 1.208427 0.001721 -0.699221
C6 1 2.412741 0.006115 0.007778
C7 1 2.406015 0.006056 1.404106
FEND
!!orient .5 .5 .5 -t30 -f17 -s0
resi TOL 21
part -1 10.5
c1 1 0 0 0
c2 1 0 0 0
c3 1 0 0 0
c4 1 0 0 0
c5 1 0 0 0
c6 1 0 0 0
c7 1 0 0 0
part 0
resi 0
!!oend
Daniel W. Paley, Columbia Nano Initiative (Columbia University), 2018. Contact the author at [email protected]
This program is licensed under the GNU General Public License and is free for anyone to use, copy, or modify. It may not be incorporated into proprietary software. The full license is included in the file license.txt.
Charlene Tsay is acknowledged gratefully for testing.