*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
I have a script called "origins.com" that will explore all allowed
origin shifts and symmetry operators to try and superimpose two
molecules based on either rmsd of identically-named atoms or overall
correlation coefficient of electron density.
It is available here:
http://bl831.als.lbl.gov/~jamesh/elves/goodies.html#origins.com
You run it like this:
origins.com right.pdb wrong.pdb P43212
The output is "neworigin.pdb" which is just a "pdbset-ed" copy of the
"wrong" PDB file that you put in, moved through the origin shift and
symmetry op that puts it closest to the "right" molecule. You need a
proper unit cell in at least one PDB header. The script requires the
CCP4 suite programs: pdbset, gensym, coordconv, sfall, and overlapmap.
This is a rather brute-force methodology (try everything, keep score),
so it is much slower than a program like EMMA, but even origins.com
doesn't usually take more than a minute to run. I suppose an
"advantage" over EMMA is that origins.com keeps chains together and
atoms are not renamed by default. (There is an auxillary output file
called newlabel_neworigin.pdb that tries to re-name atoms in the "wrong"
file after the nearest atom in the "right" file, but this feature is a
little unstable).
Phil? Is this what you were after?
Kevin? Please write me down as one vote for you to write a faster
version of this! :)
-James Holton
MAD Scientist
Phil Evans wrote:
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Is there a program somewhere which will give me the best symmetry
operation to superimpose two molecules? I'm trying to compare
different autobuilds where the different programs have built different
asymmetric units (sometimes for different fragments). The residues
numbering is random of course.
I've tried
Coot, but the writing out of symmetry-related coordinates
is currently broken,
ccp4mg but I don't seem to be able to get symmetry-related neighbours
of molecule 1 (say) around the position of molecule 2, only around its
own position
any suggestions?
Phil
