I must admit in cases like this, my first thought is - could the space group be wrong? or is it twinned - quite common in this space group.. the normal twinning tests are sometimes misleading when you have non-crystallographic translation. I would reprocess the data as P32, generate a model with two copies - presumably you already have one centred at x,y,z and the other at x,y,z+1/2, so add two centred at y,z,-z y,x,-z-1/2, plus your molecule on the two fold axis.
Then see how that works with twinned refinement - operator k,h,-l Eleanor On Thu, 26 Aug 2021 at 11:59, Schreuder, Herman /DE < herman.schreu...@sanofi.com> wrote: > Hi Peer, > > > > Normally, if one defines some residues with an occupancy below 1.0, the > nonbonding contact restraints with other residues are switched off. It is > already some time ago, but if I recall correctly, I had similar problems > that nonbonding contact restraints were not switched off for residues > related by crystallographic symmetry if they were not exactly on a > crystallographic axis. I don’t know if this is the problem that you are > facing, but here is what I did in the past: > > > > For Xplor (and I guess the same is true for Phenix), one can explicitly > switch on or off any interaction one likes, and did is what I did using > Xplor. So you might want to have a look at Phenix. For Buster, I, or > better, the global phasing people, created a gelly file to specifically > exclude troublesome restraints. > > > > Again, I don’t know if this is your problem, but if so, I would look at > Phenix or Buster to see if you can refine your structure with one of those > programs. > > > > Best, > > Herman > > > > > > *Von:* CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK> *Im Auftrag von *Peer > Mittl > *Gesendet:* Donnerstag, 26. August 2021 11:54 > *An:* CCP4BB@JISCMAIL.AC.UK > *Betreff:* [ccp4bb] chain on 2-fold axis? > > > > Der CCP4 community, > > Is there a refinement program that can handle protein monomers sitting > on crystallographic 2-folds? > > This is probably a strange question but we have the following situation. > We have a 2.6 Ang datasets in SG P3221 (Rpim=4%, Isa=19.6) and a clear > molrep solution with 2 chains, albeit with tNCS (0/0/0.5) that can be > refined to around 27/33% Rfactor. According to Vm a third chain could be > present. So far so good, but there is clear difference ED for a third > chain sitting exactly on the 2-fold. Since the protein has a peculiar > shape, one can tell even its orientation. I can relax the symmetry to > P32 (or even P1) and place the missing chain with 50% occupancy on the > 2-fold. This model can be refined, but I do not like this work around, > because the data is clearly P3221. > > Any hints on similar crystal pathologies and how they have been handled > would be helpful. > > All the best, > Peer > > ######################################################################## > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > > This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a > mailing list hosted by www.jiscmail.ac.uk, terms & conditions are > available at https://www.jiscmail.ac.uk/policyandsecurity/ > > ------------------------------ > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
