large coordinate and map shift after simulated annealing
Hi everyone, I am wondering why large coordinate shift could happen after simulated annealing (no rigid body refinement). Although it helps decreasing R values, it looks like the whole molecule shifts together by about 3 A. Is this normal? I read a post in previous phenixbb, but I still not understand it. Thank you very much! Regards, Fengyun
Hi Fengyun, Do you have a space group with a "polar axis"? For example P21? If that's the case the shift is inconsequential: shifting the entire structure arbitrarily along a polar axis does not change the calculated structure factor amplitudes. I.e. I think you don't have to worry about the shift. Ralf
Hi Ralf,
Thank you so much!
What's the meaning of polar axis?
For one structure, the space group is P1. For the other one, the space
group is C2.
Do they have polar axis?
Fengyun
引用 "Ralf W. Grosse-Kunstleve"
Hi Fengyun,
Do you have a space group with a "polar axis"? For example P21? If that's the case the shift is inconsequential: shifting the entire structure arbitrarily along a polar axis does not change the calculated structure factor amplitudes. I.e. I think you don't have to worry about the shift.
Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi Fengyun,
Thank you so much! What's the meaning of polar axis? For one structure, the space group is P1. For the other one, the space group is C2. Do they have polar axis?
C2 does, P1 is a special case: you can shift the entire structure by any amount in any direction without changing the calculated amplitudes. This paper explains it a little bit: http://cci.lbl.gov/~rwgk/my_papers/iucr/ba5051_reprint.pdf Mainly section 2.1. (We should really post-process the models after Cartesian simulated annealing to remove the arbitrary shifts...) Ralf
Hi Fengyun, interesting... Does the crystal symmetry allows an arbitrary shift along an axis? Anyway, it's unusual that the molecule shifts as a rigid body by 3A during SA, although why not ? Pavel. On 6/11/10 10:21 AM, [email protected] wrote:
Hi everyone,
I am wondering why large coordinate shift could happen after simulated annealing (no rigid body refinement). Although it helps decreasing R values, it looks like the whole molecule shifts together by about 3 A. Is this normal?
I read a post in previous phenixbb, but I still not understand it.
Thank you very much! Regards, Fengyun
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
participants (3)
-
fn1＠rice.edu -
Pavel Afonine -
Ralf W. Grosse-Kunstleve