- phenixbb - phenix-online.org

Twin refinement issues?
by Phil Jeffrey 10 Apr '09

10 Apr '09

While experimenting with twin refinement for a crystal in P321 that approximates P622 to varying extents, I noticed: | twin fraction: 0.59 twin operator: -h,-k,l | A bug, perhaps ? This twin fraction would seem to be unlikely, since twin fractions > 0.5 have no meaning. Since this data does not scale well in p622 compared to some more heavily-twinned datasets on this xtal form I think the twin fraction is certainly less than 0.5. Xtriage estimates: Statistics depending on twin laws ----------------------------------------------------------------- | Operator | type | R obs. | Britton alpha | H alpha | ML alpha | ----------------------------------------------------------------- | -h,-k,l | M | 0.150 | 0.343 | 0.347 | 0.283 | ----------------------------------------------------------------- Unless of course phenix.refine is reporting 2*alpha. Phenix v1.4-3 Intel Mac OSX 10.5.6 This particular run with: phenix.refine model-08.pdb pz7e_truncate-unique.mtz refinement.main.ncs=true strategy=individual_sites+group_adp --overwrite xray_data.r_free_flags.generate=True twin_law="-h,-k,l" This is a very early non-finessed model. Phil Jeffrey Princeton

4 6

optimize weighs
by Yusuf Akhter 09 Apr '09

09 Apr '09

Dear all, I am refining a protein structure using phenix.refine, when i tried to use optimize_wxc, optimize_wxu It always crashes and gives following error. May anybody can explain what is the problem?? Traceback (most recent call last): File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/command_line/refine.py", line 5, in <module> command_line.run(command_name="phenix.refine", args=sys.argv[1:]) File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/command_line.py", line 78, in run call_back_after_monitor_collect=call_back_after_monitor_collect) File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/driver.py", line 1308, in run log = log) File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/strategies.py", line 642, in refinement_machine log = log) File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/weight_xray_chem.py", line 327, in __init__ self.wxc_optimization() File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/weight_xray_chem.py", line 450, in wxc_optimization rw, rf, b, a, a_xc = self.wx_opt_helper(w = wxc_scale_0 * wxc, sites=True) File "/public/xtal/phenix/phenix-1.3b/phenix-1.3b/phenix/phenix/refinement/weight_xray_chem.py", line 387, in wx_opt_helper verbose = 0) TypeError: __init__() takes at least 5 non-keyword arguments (3 given) Thanks in advance. Yusuf

2 1

twin refinement error
by jonathan elegheert 08 Apr '09

08 Apr '09

Dear Phenix developers, I am rigid-body refining a PM-twinned structure (C2221 => P6x22) with the operators suggested from xtriage or SFcheck but I keep running into this same error (version 1.4.3 or latest cci aps bundles); 1/2*h-1/2*k,-3/2*h-1/2*k,-l 1/2*h+1/2*k,3/2*h-1/2*k,-l Any suggestions as to how I can resolve this? Many thanks in advance, Jonathan Elegheert PhD student Ghent University Belgium ------------------------------------------------------------------------- Traceback (most recent call last): File "/usr/local/cciapps/cci_apps_sources/phenix/phenix/command_line/refine.py", line 16, in <module> command_line.run(command_name="phenix.refine", args=sys.argv[1:]) File "/usr/local/cciapps/cci_apps_sources/phenix/phenix/refinement/command_line.py", line 89, in run call_back_after_monitor_collect=call_back_after_monitor_collect) File "/usr/local/cciapps/cci_apps_sources/phenix/phenix/refinement/driver.py", line 1121, in run call_back_after_monitor_collect = call_back_after_monitor_collect) File "/usr/local/cciapps/cci_apps_sources/phenix/phenix/refinement/strategies.py", line 190, in refinement_machine twin_law = sgtbx.rt_mx( params.twinning.twin_law ) RuntimeError: cctbx InternalError: /net/garter/scratch1/rwgk/auto_build/sources/cctbx/sgtbx/rt_mx.cpp(230): Unsuitable value for rational rotation matrix.

4 6

Phenix.refine Rigid Body multi_body_factor
by Mary Fitzgerald 08 Apr '09

08 Apr '09

I was wondering when you would want to change multi_body_factor= 1? Is this something you would change when you have multiple rigid body domains? From the name, that's what I'm guessing it's for. I'm refining a large over 400 kD multimer complex with two complexes in the ASU. So, if I have 5 chains in each multimer would I want to try to change the multi_body_factor to 10? Or does multi_body_factor affect something else. Thanks, Mary X. Fitzgerald Postdoctoral Associate Arnold Lab

2 1

optimize_wxc and wxu
by Scott Classen 08 Apr '09

08 Apr '09

Hello phenixbb, In general should one optimize wxc and wxu ONLY as a final polishing step or can this be done earlier in the refinement and then the optimized values used for subsequent rounds of refinement? If so which values in the log file should be written into the .def file and where? Here is a snippit from my def file after I optimized weights and changed the optimize flags back to False: target_weights { mode = *automatic every_macro_cycle wxc_scale = 0.5 wxu_scale = 1 wc = 1 wu = 1 fix_wxc = None fix_wxu = None optimize_wxc = False bonds_rmsd_max = 0.05 angles_rmsd_max = 3.5 optimize_wxu = False shake_sites = True shake_adp = 10 regularize_ncycles = 50 verbose = 1 wnc_scale = 0.5 wnu_scale = 1 rmsd_cutoff_for_gradient_filtering = 3 } Thanks, Scott ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Scott Classen, Ph.D. SIBYLS Beamline 12.3.1 http://bl1231.als.lbl.gov Advanced Light Source Lawrence Berkeley National Laboratory 1 Cyclotron Rd MS6R2100 Berkeley, CA 94720 C) 510.206.4418 O) 510.495.2697 Beamline) 510.495.2134 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4 3

restrains for nucleaic acid base pairs
by Imre Toeroe 08 Apr '09

08 Apr '09

Hi, Does anybody know an easy way to restrain nucleic acid base pairs (base stacking+Watson-Crick base pairing) in phenix.refine? I am refining against a 3.3 A dataset. Thanks, Imre

3 4

phenix autobuild and omit map
by rpai＠berkeley.edu 07 Apr '09

07 Apr '09

Hi, I have a loop that was improperly built when I ran autobuild. the density according to the Fobs-Fcalc maps is not that great. Someone told me that autobuild can use omit maps to build / rebuild the loops so you dont have to do it by hand. Looking at the manual, I found the command, but am a little confused. The model used is the entire model correct? the target.pdb is the chain or region that was not correctly built. For iterative building...how many cycles should I go for? phenix.autobuild data=w1.sca model=coords.pdb omit_box_pdb=target.pdb \ composite_omit_type=iterative_build_omit thanks Raj

3 2

imaginary map (anomlous difference Fourier) from phenix
by Ulrich Baumann 06 Apr '09

06 Apr '09

Hi everybody, I used phenix for calculating anomalous Fourier coefficients (script below) and I get the desired file containing columns FANOM PHANOM. When I read this into Coot for displaying the imaginary map, I only get noise, regardless of resolution cutoff etc. When I export the Fcalcs as CNS format and use CNS for the map calculation, I get very high peaks (i.e. excellent signal-to-noise ratio) on the anomalous scatterers. Inspection of the phenix fcalc file reveals that FOFCWT_no_fill equals FANOM and PHFOFCWT_no_fill is identical to PHANOM . Obviously I am doing something stupidly wrong. Setting xray_data.force_anomalous_flag_to_be_equal_to=true doesn't change anything, by the way, and i am using the latest Phenix version. I vaguely remember that there were an issue with anomlous data but can't find the thread. Thank you very much in advance for enlightening me, Ulrich # source /usr/local/phenix-1.4-3/phenix_env phenix.refine m226h1_rfree-unique.mtz m226h1_hkl2000_6_waters_001.pdb main.number_of_macro_cycles=1 \ refinement.input.xray_data.labels="F(+),SIGF(+),F(-),SIGF(-)" \ xray_data.high_resolution =3.5 strategy=none export_final_f_model=cns output.prefix=sfcalc --overwrite * Space group = 'P3121' (number 152) The mapcoeff.mtz looks like OVERALL FILE STATISTICS for resolution range 0.000 - 0.319 ======================= Col Sort Min Max Num % Mean Mean Resolution Type Column num order Missing complete abs. Low High label 1 ASC 0 49 0 100.00 23.7 23.7 88.43 1.77 H H 2 NONE 0 28 0 100.00 8.4 8.4 88.43 1.77 H K 3 NONE -69 69 0 100.00 0.6 26.2 88.43 1.77 H L 4 NONE 0.0 2220.8 0 100.00 235.55 235.55 88.43 1.77 F 2FOFCWT 5 NONE -180.0 180.0 0 100.00 2.65 89.99 88.43 1.77 P PH2FOFCWT 6 NONE 0.0 956.4 0 100.00 72.48 72.48 88.43 1.77 F FOFCWT 7 NONE -180.0 180.0 0 100.00 2.16 89.97 88.43 1.77 P PHFOFCWT 8 NONE 0.0 2202.8 117 99.84 234.51 234.51 37.16 1.77 F 2FOFCWT_no_fill 9 NONE -180.0 180.0 117 99.84 2.78 89.99 37.16 1.77 P PH2FOFCWT_no_fill 10 NONE 0.0 961.2 117 99.84 72.39 72.39 37.16 1.77 F FOFCWT_no_fill 11 NONE -180.0 180.0 117 99.84 2.19 89.95 37.16 1.77 P PHFOFCWT_no_fill 12 NONE 0.0 961.2 117 99.84 72.39 72.39 37.16 1.77 F ANOM 13 NONE -180.0 180.0 117 99.84 2.19 89.95 37.16 1.77 P PHANOM No. of reflections used in FILE STATISTICS 72570 1 1 3 88.81 -133.90 400.50 -133.90 139.90 -134.31 502.82 -134.31 502.82 -134.31 1 1 4 719.55 109.38 336.98 109.38 703.51 104.61 304.13 104.61 304.13 104.61 1 1 5 987.30 124.20 664.11 124.20 985.68 119.49 659.96 119.49 659.96 119.49 1 1 6 871.35 35.76 152.79 35.76 894.80 36.84 198.58 36.84 198.58 36.84 1 1 7 425.64 38.73 189.77 38.73 419.88 34.24 177.78 34.24 177.78 34.24 1 1 8 406.00 -154.97 53.51 25.03 416.82 -153.85 32.47 26.15 32.47 26.15 1 1 9 73.73 139.84 441.04 139.84 92.83 139.16 479.45 139.16 479.45 139.16

2 2

maps and autobuild
by chandra bose 06 Apr '09

06 Apr '09

hello, I am a first time user to phenix , and I have just completed the autobuild function,. My question may have been asked many times before, but.. The output after the cycles of autobuild/refinement gave me a refined file for a 2.2 Angstrom data set with an R/Rfree of .28/.32. I want to take a look at the density modified maps and see what I can adjust further, but am having trouble loading it in coot. Am I correct in understanding that Coot can load and construct 2f0-fc maps? if so How? : exptl_fobs_phases_freeR_flags.mtz overall_best_denmod_map_coeffs.mtz; overall_best_refine_map_coeffs.mtz; Also from reading the manual, as I understand it, the phenix.refine command will automatically generate Coot compatible maps. But if I take my autobuild.pdb file and the original mtz data file, how can I quickly obtain the same R/Rfree, what commands would I have to imput other than phenix.refine mtz pbd? Also, is it possible to make 3Fo-2Fc maps? Thanks Charlie

3 2

phenix.autobuild sa_omit map
by Jianghai Zhu 03 Apr '09

03 Apr '09

Hi, I am using phenix.autobuild to calculate a sa_omit map. The find_ncs is Auto by default. What does Auto mean here? Should I turn it off or does it matter? Thanks. -- Jianghai

2 1