Hi Peter,

here are a few considerations:

- ADP refinement: use TLS combined with group B-factor refinement (one isotropic B per residue). It may be important to optimally define TLS groups. Use NCS.

- coordinates: You may need to use custom bonds to restrain secondary structure elements. Use NCS. Note, you can define the strength of each NCS group individually. Just in case the automatic weight are not optional, try put more weight on geometry by using "wxc_scale=...". Yes, conventional rigid body refinement with small rigid groups may be good at some point, but may be too brutal as well, so constrained rigid body refinement (torsion angle dynamics) may be a good option.

- mask: sometimes, using this "optimize_mask=true" makes a big difference at such resolutions, so you may want to try.

- rigid body refinement is completely disconnected from the NCS restraints (no any restraints used during rigid body refinement).

Pavel.

On 10/15/09 1:12 PM, Peter Grey wrote:

Dear Phenix users,

I refine a large complex with three monomers in ASU at very low resolution - 4.5A - and would like to ask two questions.
- Could you give some tips about refinement at this resolution other than rigid body refinement - Can TLS and grouped B-factors work at this poor resolution ?
- I would really prefer that rigid body refinement will strictly obey the NCS that was found in the molecular replacement solution. I understand constrained NCS does not exist yet in phenix but is there something approximate that can be used in rigid body refinement ? I tried the conventional restrained NCS and the result was just like without restrains.

Thank you for your time ,

Peter.
_______________________________________________
phenixbb mailing list
[email protected]
http://www.phenix-online.org/mailman/listinfo/phenixbb
  