Eugene is away but I think the PISA symmetry checking only looks at 2 unit cells in all directions from the origin, so if the origin chain is far away you may lose contacts.
.. As someone says - always try to place the molecule reasonably close to the origin. There is an old program pdbset xyzin mol.pdb end which tells you the centre of symmetry of the mol. If it is a long way off the origin say 1.27 -0.73 0.95 an equally valid molecule could be repositioned saying pdbset xyzin mol.pdb xyzout mol-Sym.pdb symgen x-1,y+1,z-1 end Then the centre of mass would be 0.27 0.27 -0.05 You may want to split the check to do individual chains COOT will apply a shift too, once you know what it should be. Eleanor On 11 October 2017 at 18:15, Phillip Gregory Gross <phillip.gros...@uga.edu> wrote: > I got a symmetry-related molecule in the ASU (mine was about 400 A off the > origin) and ran areaimol and the PDBe PISA server, and it worked on both! > Got all the same contacts as my DIFFMODE COMPARE run. Thanks Paul! > -Phillip > > ------------------------------ > *From:* CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK> on behalf of Paul > Emsley <pems...@mrc-lmb.cam.ac.uk> > *Sent:* Wednesday, October 11, 2017 12:07:33 PM > *To:* CCP4BB@JISCMAIL.AC.UK > *Subject:* Re: [ccp4bb] areaimol missing symmetry mates > > Try moving your molecule closer to the origin, by unit cell shift or > choosing a symmetry-related molecule. > > (In Coot, you can do this by using Extensions -> Modelling -> Symm Shift > Reference Chain Here. > > Paul. >
