Yes, SA and real-space refinement are done (if enabled) for all atoms. No particular reason for this. It just needs to be fixed. Pavel On 4/6/14, 6:03 PM, Nathaniel Echols wrote:

On Sun, Apr 6, 2014 at 4:47 PM, Jorge Iulek mailto:[email protected]> wrote:

I want to test a refinement in which I fix part of my structure (say, residues 1 to 244 in chains C and D, and chains A and B completely) while the rest is to be refined with annealing. ... I thought atoms out of the selection in the output structure (chains A and B + residues 1 to 244 in chains C and D) would keep the same coordinates as they were input. But I see all structure atoms changed coordinates. I also tried to turn off tls and ncs usage, but the picture is the same. What do I miss here, to be able to fix some atoms during refinement?

I think the annealing may only work with the entire model and ignore the atom selection completely.

-Nat

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

Dear Eckhard, Yes, good advice, to use harmonic restraints does enough to what I wanted. Thanks. Nevertheless, I have not managed to use the model restraints yet, but I will study it better. Yours, Jorge On 04/07/2014 06:04 PM, Eckhard Hofmann wrote:

But could you not just switch on tight reference model restraints or harmonic restraints for this part of your structure? Of course this is not leaving them untouched, but if you set sigma to 0.01, then they should pretty much stay the same. Eckhard (In main: set reference_model_restraints = True and adjust parameters in section reference_model)

Am 07.04.2014 05:56, schrieb Pavel Afonine:

...

Yes, SA and real-space refinement are done (if enabled) for all atoms. No particular reason for this. It just needs to be fixed.

Pavel

On 4/6/14, 6:03 PM, Nathaniel Echols wrote:

...

On Sun, Apr 6, 2014 at 4:47 PM, Jorge Iulek mailto:[email protected]> wrote:

I want to test a refinement in which I fix part of my structure (say, residues 1 to 244 in chains C and D, and chains A and B completely) while the rest is to be refined with annealing. ... I thought atoms out of the selection in the output structure (chains A and B + residues 1 to 244 in chains C and D) would keep the same coordinates as they were input. But I see all structure atoms changed coordinates. I also tried to turn off tls and ncs usage, but the picture is the same. What do I miss here, to be able to fix some atoms during refinement?

I think the annealing may only work with the entire model and ignore the atom selection completely.

-Nat

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

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