[buster-discuss] BUSTER refinement - constrained occupancy groups?

[Clemens Vonrhein]

Dear Jose,

On Wed, Apr 20, 2016 at 05:01:37PM +0100, Jose Artur Brito wrote:
> Dear All,
> is it possible to create "constrained occupancy groups" similarly to
> what phenix.refine has?

I don't know what phenix.refine has - but BUSTER has this for some
time now. See e.g.

  Moniot, S., Bruno, S., Vonrhein, C., Didierjean, C., Boschi-Muller,
  S., Vas, M., ... & Corbier, C. (2008). Trapping of the
  Thioacylglyceraldehyde-3-phosphate Dehydrogenase Intermediate from
  Bacillus stearothermophilus DIRECT EVIDENCE FOR A FLIP-FLOP
  MECHANISM. Journal of Biological Chemistry, 283(31), 21693-21702.

where we used an occupancy refinement scheme that sounds similar to
what you are trying to achieve.

> I have a heme c with alternate conformation, a cysteine, also with
> alternate conformation, and an active site ligand. I want to
> constrain the occupancy of (heme + ligand + cysteine - alternate
> conformations A), to one group and the occupancy of (heme + cysteine
> - alternate conformations B), to another.

Unless I understand your situation completely wrong, the easiest
approach seems to be: ensure your PDB file contains

  heme.A ligand.A cysteine.A = all at same occupancy (e.g. 0.9)
  heme.B ligand.B cysteine.B = all at same occupancy (e.g. 0.1)

This can be done in Coot or via old-fashioned editing. Then run

  pdb2occ -p your.pdb -o your.occ

and edit the resulting your.occ file according to:

 (1) all *.A atoms should be added to the same group

 (2) all *.B atoms should be added to the same (different) group

 (3) you can restrain the sum of occupancies to be 1 (optional)

This should then look roughly like that (assuming heme=A1001,
ligand=A1002 and cysteine = A123):

  NOTE BUSTER_RESET_CONSTANT_COMBINE

  NOTE BUSTER_SET AltOccAll = Empty

  NOTE ----- first group:
  NOTE BUSTER_SET AltOcc1    = { A|1001:*.A}
  NOTE BUSTER_SET AltOcc1    = { A|2001:*.A} + AltOcc1
  NOTE BUSTER_SET AltOcc1    = { A|123:*.A}  + AltOcc1
  NOTE BUSTER_SET AltOccAll = AltOccAll + AltOcc1   
  NOTE BUSTER_COMBINE OCC AltOcc1

  NOTE ----- second group:
  NOTE BUSTER_SET AltOcc2    = { A|1001:*.B}
  NOTE BUSTER_SET AltOcc2    = { A|2001:*.B} + AltOcc2
  NOTE BUSTER_SET AltOcc2    = { A|123:*.B}  + AltOcc2
  NOTE BUSTER_SET AltOccAll = AltOccAll + AltOcc2
  NOTE BUSTER_COMBINE OCC AltOcc2

  NOTE ----- restraint sum(occ) to one:
  NOTE BUSTER_OCCSUM 1.0 0.005 A|123:CA.A A|123:CA.B

  NOTE BUSTER_SET FixOcc = All
  NOTE BUSTER_SET FixOcc = FixOcc \ AltOccAll
  NOTE BUSTER_CONSTANT OCC FixOcc

See also

  https://www.globalphasing.com/buster/wiki/index.cgi?AutoBusterExampleOccupancyRef
  https://www.globalphasing.com/buster/wiki/index.cgi?AutoBusterLigandAndOccRef
  https://www.globalphasing.com/buster/manual/gelly/manual/gelly3.html

for some examples and the full details.

If you know of a recent example in the PDB that shows a similar
pattern, please let us know (we might work that into an online
tutorial).

Cheers

Clemens

-- 

***************************************************************
* Clemens Vonrhein, Ph.D.     vonrhein AT GlobalPhasing DOT com
*
*  Global Phasing Ltd.
*  Sheraton House, Castle Park 
*  Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group      (http://www.globalphasing.com)
***************************************************************