Twin fraction in phenix refine
Dear all, phenix.xtriage estimates a twin fraction of between 24 and 30% but then when I start the refinement I get 47-49%. In CNS if I fix the twin fraction in a series of refinements the Rfactors jump from 44% (twin fraction 0) to 37% (from 25% to 49%, only marginal improvement upon going from alpha=25% to 49%). Questions: 1. Is the refinement of alpha dependent on the model? 2. I am tempted to work with the 30% that xtriage estimates but do not know how: is there a way to fix alpha? Thanks Pietro -- Sent from my Desktop Pietro Roversi EP Abraham Fellow in Biochemistry - Lincoln College - Oxford Sir William Dunn School of Pathology, Oxford University South Parks Road, Oxford OX1 3RE, England UK Tel. 0044-1865-275385
Hi Pietro, 1. Is the refinement of alpha dependent on the model?
yes
2. I am tempted to work with the 30% that xtriage estimates but do not know how: is there a way to fix alpha?
no. not yet. P
Thanks
Pietro
-- Sent from my Desktop
Pietro Roversi EP Abraham Fellow in Biochemistry - Lincoln College - Oxford Sir William Dunn School of Pathology, Oxford University South Parks Road, Oxford OX1 3RE, England UK Tel. 0044-1865-275385
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
-- ----------------------------------------------------------------- P.H. Zwart Beamline Scientist Berkeley Center for Structural Biology Lawrence Berkeley National Laboratories 1 Cyclotron Road, Berkeley, CA-94703, USA Cell: 510 289 9246 BCSB: http://bcsb.als.lbl.gov PHENIX: http://www.phenix-online.org CCTBX: http://cctbx.sf.net -----------------------------------------------------------------
Hi Pietro,
phenix.xtriage estimates a twin fraction of between 24 and 30% but then when I start the refinement I get 47-49%.
- did you use phenix.refine for refinement? - are you using the latest (or one of) version of PHENIX?
2. I am tempted to work with the 30% that xtriage estimates but do not know how: is there a way to fix alpha?
The benefits of fixing twin fraction are not clear to me. Refinement of it should always be better. If it's not the case I would be interested to have a closer look at this problem. Pavel.
Hi, I was trying to refine a model using a fake mtz file created from model. This mtz file contains h, k, l, FMODEL, PHIFMODEL I used in the command line as follows: phenix.refine model.updated.pdb_modified.pdb elbow.model_updated_pdb_modified_pdb.all.cif model.updated.pdb.mtz input.xray_data.labels=FMODEL strategy=individual_sites main.bulk_solvent_and_scale=false main.number_of_macro_cycles=10 I got the following error iotbx.reflection_file_utils.Sorry_Not_a_suitable_array: Not a suitable array of observed xray data: refinement.input.xray_data.labels=FMODEL No array of observed xray data found. I think the command I used is not correct. I need to know the correct command to run this refinement. Thanks... Raja
Hi Raja, you should make sure that the fake fobs are real and not complex values. I presume you used phenix.pdbtools to create fake fobs, then the example of a correct command is: phenix.pdbtools model.pdb --f-model type=real label=f-obs r_free=0.1 high_res=1.5 Note, the command above gives you Fcalc computed from atomic model: Fmodel = scale * exp(-h*U_overall*ht) * (Fcalc + k_sol * exp(-B_sol*s^2) * Fmask) You can also add bulk solvent and anisotropy contributions to have full Fmodel: phenix.pdbtools pdb1akg.ent --f-model type=real label=f-obs r_free=0.1 high_res=1.5 k_sol=0.35 b_sol=50.0 b_cart="1 2 3 4 5 6" The phenix.refine command below seems to be correct. Pavel. On 1/13/10 10:39 AM, Raja Dey wrote:
Hi, I was trying to refine a model using a fake mtz file created from model. This mtz file contains h, k, l, FMODEL, PHIFMODEL
I used in the command line as follows: phenix.refine model.updated.pdb_modified.pdb elbow.model_updated_pdb_modified_pdb.all.cif model.updated.pdb.mtz input.xray_data.labels=FMODEL strategy=individual_sites main.bulk_solvent_and_scale=false main.number_of_macro_cycles=10
I got the following error
iotbx.reflection_file_utils.Sorry_Not_a_suitable_array: Not a suitable array of observed xray data: refinement.input.xray_data.labels=FMODEL No array of observed xray data found.
I think the command I used is not correct. I need to know the correct command to run this refinement. Thanks...
Raja _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi, In phenix.refine can the shake option be applied to only a portion of the molecule? (for example to side-chains only). Thanks, Esmael Haddadian
Hi Esmael, yes. Example: phenix.refine model.pdb data.mtz modify_start_model.sites.shake=0.5 modify_start_model.selection="sidechain" Pavel. On 1/13/10 4:06 PM, esmael wrote:
Hi,
In phenix.refine can the shake option be applied to only a portion of the molecule? (for example to side-chains only).
Thanks,
Esmael Haddadian
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi Pavel, Thanks, could simulated annealing also be applied only to a portion of the molecule? Esmael On Jan 13, 2010, at 6:22 PM, Pavel Afonine wrote:
Hi Esmael,
yes. Example:
phenix.refine model.pdb data.mtz modify_start_model.sites.shake=0.5 modify_start_model.selection="sidechain"
Pavel.
On 1/13/10 4:06 PM, esmael wrote:
Hi,
In phenix.refine can the shake option be applied to only a portion of the molecule? (for example to side-chains only).
Thanks,
Esmael Haddadian
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi Esmael, no, SA is always applied to the whole structure. It is in to-to list to make it possible to apply SA to a part of a structure. Pavel. On 1/13/10 4:28 PM, esmael wrote:
Hi Pavel,
Thanks, could simulated annealing also be applied only to a portion of the molecule?
Esmael
On Jan 13, 2010, at 6:22 PM, Pavel Afonine wrote:
Hi Esmael,
yes. Example:
phenix.refine model.pdb data.mtz modify_start_model.sites.shake=0.5 modify_start_model.selection="sidechain"
Pavel.
On 1/13/10 4:06 PM, esmael wrote:
Hi,
In phenix.refine can the shake option be applied to only a portion of the molecule? (for example to side-chains only).
Thanks,
Esmael Haddadian
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
Hi all, What would be the best strategy to only refine the side-chains in a give structure with minimal movement of the backbone? Thanks, Esmael
Hi Esmael, yes, you can do it in phenix.refine. In general, you can refine coordinates, b-factors or occupancies of any selected atoms with any available strategy (SA is an exception). Example: phenix.refine mode.pdb data.mtz refine.sites.individual="sidechain" will refine the coordinates of sidechain atoms only, and B-factors of all the atoms. There are more examples here: http://www.phenix-online.org/documentation/refinement.htm including syntax for atom selections. Let me know if you have any other questions or problems. Pavel. On 1/17/10 7:32 PM, Esmael Haddadian wrote:
Hi all,
What would be the best strategy to only refine the side-chains in a give structure with minimal movement of the backbone?
Thanks, Esmael
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
i am solving one of my structure with resolution 3.0 . i used phenix auto build for building . R factor and Rfree is now 20.4 24.9 but 69 residue out of 800 residue are outlier in ramchandran plot . further refinment increses the r factor how should i can step further thanx
Hello, 1) try playing with the target weight (refinement of coordinates). For this, try an array of values for wxc_scale and see which one works best; 2) manually fix the outliers and do "1)". Pavel. On 1/14/10 3:51 AM, [email protected] wrote:
i am solving one of my structure with resolution 3.0 . i used phenix auto build for building . R factor and Rfree is now 20.4 24.9 but 69 residue out of 800 residue are outlier in ramchandran plot . further refinment increses the r factor how should i can step further thanx
------------------------------------------------------------------------
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
On Jan 14, 2010, at 3:51 AM, [email protected] wrote:
i am solving one of my structure with resolution 3.0 . i used phenix auto build for building . R factor and Rfree is now 20.4 24.9 but 69 residue out of 800 residue are outlier in ramchandran plot . further refinment increses the r factor how should i can step further
In addition to what Pavel said: 20.4/24.9 is excellent for 3.0A resolution (try phenix.polygon to see for yourself) - and perhaps unrealistic, since you have so many Ramachandran outliers. If you fix the outliers (and any other geometry problems) and R-free increases again after refinement, you may still want to accept those results, if the final values of R/R-free are otherwise reasonable. -Nat ------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]
participants (8)
-
esmael -
Esmael Haddadian
-
Nathaniel Echols -
Pavel Afonine -
Peter Zwart -
Pietro Roversi -
Raja Dey -
v_kukshal＠cdri.res.in