Hi Tim,
I'm all for automation and 'smart' software but it should not be a substitute
for common sense. In the case of occupancies, not all software seems to be
smart. There are about 1100 structures in the PDB that have problems with wrong
occupancies at special positions and a good part of them seem to be fairly new
(pdbids 3[a-n]??).
I agree that Jackie should build the whole molecule to make sure all restraints
kick in. It's only a little extra work to set the occupancy correctly for all
atoms. In Coot one can do that with the 'residue info' option.
Cheers,
Robbie Joosten
> Date: Sat, 16 Oct 2010 10:25:47 +0200
> From: t...@shelx.uni-ac.gwdg.de
> Subject: Re: [ccp4bb] molecule on symmetry axis
> To: CCP4BB@JISCMAIL.AC.UK
>
> Hello Jackie Vitali,
>
> What happens if you model the tartaric acid as if it were not on a symmetry
> axis
> and then let phenix deal with the symmetry? I once had a water of which refmac
> stubbornly lowered the occupancy to 0.5 even though i kept on bringing it back
> to 1.0 until I realised that it was on a 2-fold axis. And this was seven years
> ago, so by now programs might be able to deal with more complicated molecules
> than just an oxygen.
>
> As for the SO4, you should set the occupancy of S and O to 0.33 because you
> only
> have one third per asymmetric unit. Now since the other oxygens belong to the
> same molecule, their occupancies should also be 0.33. But again, of you leave
> it
> at 1 and feed it into phenix, that should be corrected for. If it's not, try
> refmac!
>
> Cheers, Tim
>
> On Fri, Oct 15, 2010 at 06:28:52PM -0400, Jacqueline Vitali wrote:
> > Dear colleagues,
> >
> > I have a tartaric acid on a two fold axis with its two halves related
> > by the two fold. How do I refine this in Phenix?
> >
> > Also I have a SO4 on a 3 fold with S and one O on tthe 3 fold. The
> > other 3 oxygens are related by the 3-fold. How do I refine this in
> > phenix? I can put S and one O occupancy 1, what occupancy do I put
> > for the 3 oxygens that overlap their symmetry mates?
> >
> > And how do I maintain stereochemistry around the symmetry axis? These
> > are not just one atom. For the tartaric acid the 2 fold goes through
> > the middle of the bond. I could split it in two halves but I do not
> > see how to keep stereochemistry.
> >
> > I would appreciate all suggestions.
> >
> > I apologize because the question should go to another newsgroup but I am
> > still working with my subscription in phenixbb.
> >
> > Jackie Vitali
>
> --
> --
> Tim Gruene
> Institut fuer anorganische Chemie
> Tammannstr. 4
> D-37077 Goettingen
>
> phone: +49 (0)551 39 22149
>
> GPG Key ID = A46BEE1A
>
