- phenixbb - phenix-online.org

nightly builds nit (and a question)
by Ben Eisenbraun 21 Apr '09

21 Apr '09

Howdy, I'm not sure you are soliciting bugs on the nightly builds, but I thought I'd mention these. Please let me know if you don't need/want them. I downloaded and installed the 1.4-37 nightly builds. The installations went fine for linux and OS X PPC/x86. There are some absolute symlinks in the installation that don't work: phenix-1.4-37/build/intel-linux-2.6/base/bin/wx-config -> /net/longnose/scratch1/phenix/phenix-1.4-37/build/intel-linux-2.6/base/lib/wx/config/gtk2-ansi-release-2.8 On the OS X there are some that don't appear to matter, e.g.: phenix-1.4-37/build/mac-intel-osx/base/bin/MacPyMOL.app/pymol/test/pdb -> /Volumes/Macintosh HD/Users/delwarl/pdb/entries And some that maybe do: phenix-1.4-37/build/mac-intel-osx/base/bin/python2.6 -> /net/patchnose/scratch1/phenix/phenix-1.4-37/build/mac-intel-osx/base/Python.framework/Versions/2.6/bin/python2.6 Since the nightly builds don't build the ext_ref_files, should I just symlink them to the ones in the 1.4-3 release? Thanks. -ben -- | Ben Eisenbraun | Software Gunky | | Structural Biology Grid | http://sbgrid.org | | Harvard Medical School | http://hms.harvard.edu |

2 1

SA_OMIT map at low resolution
by 邬金 21 Apr '09

21 Apr '09

Hi, I would like to calculate sa_omit map for a low resolution structure (4.0 Ang) to validate the conformation in some regions. What's the proper way to do it? Followed the instruction, I tried: phenix.autobuild data=mydata.mtz model=mybestmodel.pdb omit_box_pdb=mybestmodel.pdb \ omit_res_start_list=100 omit_res_end_list=200 omit_chain_list=L \ composite_omit_type=sa_omit I have several questions: 1. "omit_box_pdb=mybestmodel.pdb": If I define the region that I want to omit, do I need to define the omit_box pdb as well? Should it be complete model or the omit region? 2. I have 4 fold NCS axis, if I want to omit a loop in four copies, how can I define 4 corresponding regions in different chains? e.g omit region: 100-200 in chain L, M, N, O: omit_chain_list = L omit_chain_list = M omit_chain_list = N omit_chain_list = O omit_res_end_list = "200" omit_res_start_list = "100" 3. If I want to omit several loops in one copy, should I turn off NCS? Like: find_ncs=False input_ncs_file= None optimize_ncs=False 4. How to define several omit regions in one chain? Many thanks, Jin _________________________________________________________________ 上Windows Live 中国首页，下载最新版 MSN！ http://im.live.cn/

2 1

Re: [phenixbb] phenixbb Digest, Vol 41, Issue 12
by Ho-Leung Ng 21 Apr '09

21 Apr '09

Hi Zach, You probably have the wrong space group. P3 is rarely represented in the PDB, with only 67 deposited structures out of 57000. You can use Phaser to automatically try all the other P3 related space groups. phenix.xtriage will also try to predict the space group for you. Good luck! Ho UC Berkeley

4 5

phenix.automr high zscore but too many clashes
by zach powers 18 Apr '09

18 Apr '09

Hi, I have a problem and i wonder if someone has had a similar experience. I have been using phenix.automr program to find an MR solution for a protein. When I ask Phaser to use copies=2 I find a number of solutions with poor z-scores (3-4). When I ask it to use copies=3, I get great Z-scores (see below) but no solution due to the high number of clashes (>100!). As an x-ray newbie I am a bit perplexed of what to make of this: the solution is not good because there is not enough space in the unit cell to comfortably accommodate all three molecules yet the Z-score indicates the structure is good. My structure does have several loops and these may be contributing to the clashing residues. As a newbie I have a newbie question - what does this mean (great z-score but no solutions)? Is this non-solution a possible solution if I play with it, or is the packing simply too tight and the Z-scores are not valid? In the meantime I am chopping my protein into sub-domains as recommended in the Phaser documentation, but if anyone has seen something like this before, any suggestions are welcome. thanks zach charlop-powers Packing Table: Space Group P 3 ------------------------------ Solutions accepted if number of clashes = 135 (lowest number of clashes in list) provided this number of clashes <= 10 (maximum number of allowed clashes) # #Clashes # Accepted Annotation 1 188 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=10.4 2 156 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=10.2 3 151 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=9.8 4 146 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=9.7 5 187 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 TFZ=10.5 6 202 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=9.2 7 170 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=9.1 8 144 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 TFZ=10.2 9 177 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=9.1 10 153 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.9 11 150 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.8 12 217 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.6 13 182 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.6 14 135 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.5 15 175 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.3 16 159 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.3 17 192 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.3 18 165 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 TFZ=9.2 19 169 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 TFZ=9.1 20 158 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.0 21 173 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=8.0 22 154 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=16.3 TFZ=8.9 23 206 NO RFZ=20.3 TFZ=18.2 PAK=10 LLG=431 RFZ=18.5 TFZ=7.8 0 accepted of 23 solutions

4 4

phenix.autobuild error
by Andreas Förster 15 Apr '09

15 Apr '09

Hey all, I'm calling phenix.autobuild with the following command: phenix.autobuild data=second_try_33.1.mtz model=second_try_33.1.pdb \ > seq_file=name_139_414.seq ps_in_rebuild=True On my MacBook Pro (OSX 5.6, phenix Version: 1.4, Release tag: 3, cctbx tag: 2008_12_07_1353, Platform: mac-intel-osx osx-10.5.6), I get the following error: Exception RuntimeError: 'maximum recursion depth exceeded while calling a Python object' in <type 'exceptions.KeyError'> ignored 108409888 On a linux box (RHEL 5.2, phenix Version: 1.4, Release tag: 3, cctbx tag: 2008_12_07_1353, Platform: intel-linux-2.6 redhat-e5.2) I get this error: ******************************************************************************** Failed to carry out AutoBuild_build_cycle: child process stderr output: command: 'csh /csb/home/andreas/name/P6big/second_try/autobuild/AutoBuild_run_2_/TEMP0/RUN_FILE_1' tset: standard error: Invalid argument ******************************************************************************** Each time, the script was run on the same set of files residing on the linux box. I either ssh into the machine and run phenix.autobuild in an ssh shell or mount my linux home directory via MacFusion and run phenix.autobuild on the Mac. I haven't tried copying the files to the Mac, but other phenix commands work with the remote data (e.g. phenix.autobuild data=data.mtz solvent_fraction=.6 ps_in_rebuild=True model=coords.pdb maps_only=True) What's the problem? Andreas -- Andreas Förster, Research Associate Paul Freemont & Xiaodong Zhang Labs Department of Biochemistry, Imperial College London http://www.msf.bio.ic.ac.uk

2 1

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