Hi Nat,

the way it's implemented currently: you can't add constraints on top of another constraints. I'm not aware of any existing program that can do it, although technically it's easy to implement.

However, Dalibor can do what he wants (if I understand two previous emails correctly). In fact, phenix.refine will do it automatically, so there is no need to even do anything special.

Pavel.



On 11/25/10 11:08 AM, Nathaniel Echols wrote:
On Thu, Nov 25, 2010 at 10:26 AM, Pavel Afonine <[email protected]> wrote:
In the example below I see residue #30 has two conformations, A and B, and they are automatically constrained-refined in phenix.refine and their sum adds up to 1 (0.52+0.48).
Same for residue #451.

Although I see that occupancy of "AGLU A 30 " = occupancy of "AALA A 451 ", it is not guaranteed in refinement, since otherwise that would be a double-constrained refinement:

constraint #1: occupancy(AGLU A 30) + occupancy(BGLU A 30)=1
constraint #2: occupancy(AGLU A 30) = occupancy(AALA A 451)

which is not currently available.

Wait, now I'm confused too - isn't this the entire point of the constrained_group setting?  For example, the parameters below:

refinement.refine.occupancies.constrained_group {
  selection = "chain A and resseq 30"
  selection = "chain A and resseq 451"
}

If both selections have alternate conformers A and B, and the occupancies for A and B are both 0.5, what would phenix.refine do?

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