question is after the setup here : i have a ligand with multiple conformers. i gave each conformer a different chain label. refining occupancy for the overlapped parts kicks that part out of the density. i have remove_selection for the parts that are 1.0 and they seem OK. a single conformer stays in the density. i tried starting with and without site or occupancy refinement, no help. i do not want group occupancy refinement b/c some parts do not overlap - i.e. they "should" be occupancy 1.0 -- i tried it anyways with constrained_group =... to check. is the chain label an improper way to do this? -bryan p.s: Happy New Year with Blue Moon Eclipse (~10degW->180degE long. eastern HS)!
i do not want group occupancy refinement b/c some parts do not overlap - i.e. they "should" be occupancy 1.0 -- i tried it anyways with constrained_group =... to check.
is the chain label an improper way to do this?
Yes, column 17 "altloc" is for this purpose. For example: ATOM 111 CB AVAL A 8 4.063 1.695 12.923 0.70 3.19 C ATOM 112 CB BVAL A 8 4.269 1.433 13.215 0.30 9.52 C ^ altloc The rules for nonbonded interactions are: " " interacts with everything "A" interacts with " " and "A" only "B" interacts with " " and "B" only Ralf
Bryan,
Are you saying that your ligand can adopt two completely separate conformers, or that part of it moves around? Either way, you do want grouped refinement with the altloc, either on the whole or part of the ligand that is different. Altloc automatically performs group refinement. It doesn't make physical sense to do otherwise, as you are essentially assigning a percent that the ligand is in the different conformers, rather than saying that part of the molecule is only there part of the time.
Happy New Year,
Kendall Nettles
On 12/31/09 5:41 PM, "Bryan W. Lepore"
Hi Bryan, I'm not sure I understand what exactly you are trying to do (at least at this time), but if you send me your PDB file and tell me what exactly should happen to its content during refinement, then I will most likely send you an example of how to do it. The past experience shows that this kind of questions is usually easy-fixable. Happy New Year! Pavel. On 12/31/09 2:41 PM, Bryan W. Lepore wrote:
question is after the setup here :
i have a ligand with multiple conformers. i gave each conformer a different chain label. refining occupancy for the overlapped parts kicks that part out of the density. i have remove_selection for the parts that are 1.0 and they seem OK. a single conformer stays in the density. i tried starting with and without site or occupancy refinement, no help.
i do not want group occupancy refinement b/c some parts do not overlap - i.e. they "should" be occupancy 1.0 -- i tried it anyways with constrained_group =... to check.
is the chain label an improper way to do this?
-bryan
p.s: Happy New Year with Blue Moon Eclipse (~10degW->180degE long. eastern HS)!
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
participants (4)
-
Bryan W. Lepore -
Kendall Nettles -
Pavel Afonine -
Ralf W. Grosse-Kunstleve