- phenixbb - phenix-online.org

coot from phenix
by Andreas Förster 24 Oct '11

24 Oct '11

Dear all, I'm running phenix 1.7.2-869 on RHEL 6.1 64-bit. When I click on "Open in Coot" after refinement, coot doesn't open and I get the following error in the shell from which I started phenix: /csb/soft/Linux64/share/phenix-1.7.2-869/phenix-1.7.2-869/phenix/phenix/command_line/start_coot.sh: line 33: /csb/soft/Linux/src/coot-0.6/bin/coot: No such file or directory In the script that launches coot, $PHENIX_COOT is defined as /csb/soft/Linux/src/coot-0.6/bin/coot, which doesn't exist (and is a path that we used in the old days of 32-bit). Typing phenix.find_coot_command in the shell from which phenix was started returns /csb/soft/Linux64/src/coot-0.7-rev3689/bin/coot, which is the right coot. To me it seems the phenix script to launch coot is falling back on an internally defined $PHENIX_COOT variable (that's not defined from the shell). That's shouldn't be, should it? How do I fix this short of hardwiring $PHENIX_COOT as /csb/soft/Linux64/src/coot-0.7-rev3689/bin/coot? Thanks. Andreas -- Andreas Förster, Research Associate Paul Freemont & Xiaodong Zhang Labs Department of Biochemistry, Imperial College London http://www.msf.bio.ic.ac.uk

2 2

Unusual CCP4 to Phenix data conversion result
by R.M. Garavito 24 Oct '11

24 Oct '11

In upgrading to CCP4 6.2.0 and Phenix-1.7.2-869 (from CCP4 6.1.13 and Phenix-1.7.1-743, respectively) a bug has crept into reading of the MTZ files. We tend to scale and merge XDS processed data with Scala in CCP4, then use the resulting data in CCP4 and Phenix (with great success I might add). However, I just noticed that in testing the upgraded CCP4 and Phenix on OSX 7.2 (Lion) that the MTZ file is no longer read correctly. An example is below where after the output from scaling and merging of nonanomalous data from the CCP4i GUI gives, at the end of FREERFLAG, some 47,000 reflections. Putting this into xtriage or the reflection file editor in the Phenix GUI, one finds that the data for the F and the DANO columns are used; just selecting Fs alone is impossible. The end result is a doubling of the reflection number and the switching of the anomalous flag on. Processing with earlier versions did not do this except when actually using anomalous data. Why this has happened is not clear to me, but the column order in the MTZ file changed slightly between CCP4 6.1.13 and CCP4 6.2.0. Any ideas, am I missing something, or is this a push to only refine with intensities? Regards, Michael ***************** From Scala (3.3.20) Total number unique : 46813 From FREERFLAG Number of Reflections = 47044 (including non-observed NaN reflections) MTZ column labels H K L FreeR_flag F_S13E_X2 SIGF_S13E_X2 DANO_S13E_X2 SIGDANO_S13E_X2 F_S13E_X2(+) SIGF_S13E_X2(+) F_S13E_X2(-) SIGF_S13E_X2(-) ISYM_S13E_X2 IMEAN_S13E_X2 SIGIMEAN_S13E_X2 I_S13E_X2(+) SIGI_S13E_X2(+) I_S13E_X2(-) SIGI_S13E_X2(-) ****************** In Phenix: Xtriage output choosing "Imean" Miller array info: /Users/garavito/ccp4_projects/sus1_S13/reflections_tmp.mtz:IMEAN_S13E_X2,SIGIMEAN_S13E_X2 Observation type: xray.amplitude Type of data: double, size=46772 Type of sigmas: double, size=46772 Number of Miller indices: 46772 Anomalous flag: False Xtriage output choosing "Fs" Miller array info: /Users/garavito/ccp4_projects/sus1_S13/reflections_tmp.mtz:F_S13E_X2,SIGF_S13E_X2,DANO_S13E_X2,SIGDANO_S13E_X2 Observation type: xray.amplitude Type of data: double, size=88486 Type of sigmas: double, size=88486 Number of Miller indices: 88486 Anomalous flag: True **************************************************************** R. Michael Garavito, Ph.D. Professor of Biochemistry & Molecular Biology 513 Biochemistry Bldg. Michigan State University East Lansing, MI 48824-1319 Office: (517) 355-9724 Lab: (517) 353-9125 FAX: (517) 353-9334 Email: garavito(a)msu.edu ****************************************************************

3 6

phenix.pdbtools
by Abhinav Kumar 21 Oct '11

21 Oct '11

Hi, I had been using the pdbtools to obtain the rmsd on various geometry parameters using the command: phenix.pdbtools my.pdb --show-geometry-statistics Now pdbtools does not recognize the option "--show-geometry-statistics" What is the option now? I am using phenix-1.7.2-869. -- Thanks, Abhinav JCSG@SSRL, SLAC Phone: (650) 926-2992 Fax: (650) 926-3292

3 3

ligands at a special position
by Subhani Bandara 19 Oct '11

19 Oct '11

Hi, I am trying to solve a structure of a protein that is cocrystallized with a chelator complex using phenix (GUI). The complex is Tb(DPA)3 and DPA stands for dipicolinic acid. In the structure (C222) there are few Tb complexes on the two fold axis with symmetry coordinates 1/4X, 1/4Y, Z. In those, Tb and the center C (C4) and N (N1) of one DPA are sitting directly on the 2 fold axis. So I left only half of that DPA ligand structure on PDB file and changed the occupancy of N1 and C4 to 1/2 while keeping the occupancy of all other atoms of the half ligand structure at 1.00. Then I refined it while keeping above occupancies and xyz coordinated of the ligand fixed. After doing this the other half of the ligand could be built by symmetry, but it left a space between two halves(so it was not as one ligand). Can you please let me know what I can do to fix this structure. Also is there a way to keep only x and y coordinates fixed while varying z during refinement. Thanks Subhani

3 4

Re: [phenixbb] Phenix building residues not found in input sequence file
by Thomas C. Terwilliger 18 Oct '11

18 Oct '11

Hi Allan, I assume that there is density showing protein in the region where autobuild is building these extra residues...is that right? If so...then is the remainder of the model matched very well to sequence (i.e., is more or less what you expected to see found, plus there is extra density?) It could be that there is really something in the crystal that you didn't expect. It could also be that part of the structure is not (or not correctly) assigned to sequence, so that this part of the structure really does go with part of your sequence. Does that help? It would be interesting to hear more details of this. All the best, Tom T >> Hi everyone, >> >> Apparently my Phenix (1.6.3-473) is building in extra residues in the >> C-terminal region, which is not found in my input sequence file. >> >> It's trying to build a model which is longer than I expected. Phenix >> building 87 extra amino acids. Any clue what's wrong? >> >> I made sure that these amino acids are not residues found in the >> N-terminal region or anywhere else in the protein. >> >> Allan >> >> -- >> Allan Pang >> >> PhD Student >> >> G35 Joseph Priestley Building >> Queen Mary University of London >> London >> E1 4NS >> >> Phone number: 02078828480 >>

1 0

post-doc position at SSRL
by van den Bedem, Henry 18 Oct '11

18 Oct '11

Modeling Conformational Ensembles of Proteins and Complexes A post-doctoral position with a focus on modeling conformational ensembles of proteins and complexes is available immediately in the Structural Genomics Division of Stanford Synchrotron Radiation Lightsource at the SLAC National Accelerator Laboratory (http://ssrl.slac.stanford.edu). Macromolecules and assemblies are studied by a variety of experimental techniques, including protein crystallography and Small Angle X-ray Scattering. Nano-crystallography and single particle imaging data from the Linac Coherent Light Source (LCLS), a first-of-its-kind free electron laser, are also being targeted at these systems. Accurately modeling the conformational variability of macromolecules is essential to understanding their cellular function. This position at the forefront of computational structural biology aims to develop advanced algorithms to fit domains or proteins to experimental data from multiple sources while allowing for significant conformational changes. The goal is to automatically compute 3-D models, and to provide structural insights into functionally relevant dynamics, especially for time-resolved experiments or a series of static structures. The position offers the opportunity to closely collaborate with the Stanford Computer Science Department and work with X-ray, SAXS and LCLS nanocrystallography data. SLAC houses ample computational resources. Funding is available for up to 3 years. The starting date is as soon as possible. Candidates should have a recent PhD in computer science, applied mathematics, computational biology/crystallography or a related field with strong programming skills. Please send a CV, a short summary of your research experience and two letters of recommendation to Henry van den Bedem (vdbedem(a)slac.stanford.edu). Further details are available upon request. Relevant publications: Yao et al. Sampling-based exploration of folded state of a protein under kinematic and geometric constraints. Proteins: Structure, Function, and Bioinformatics (in press). Fraser et al. (2011) Accessing protein conformational ensembles using room-temperature X-ray crystallography. PNAS 108:16247-16252 van den Bedem et al. (2009) Modeling discrete heterogeneity in X-ray diffraction data by fitting multi-conformers. Acta Cryst. D65:1107-1117

1 0

Re: [phenixbb] STOPWIZARD
by Thomas C. Terwilliger 17 Oct '11

17 Oct '11

Hi Morten, Yes, you can use STOPWIZARD to stop a phenix wizard, including mr_rosetta. However the "restarting" part is no longer enabled for most of the wizards. For mr_rosetta, you can restart. Here is how: You can re-start mr_rosetta only at the beginning of major stages (like "place_model", "rosetta_rebuild" etc)...but not in between. Normally at the end of a major stage a .pkl file is written out with text like "type this to see all the results". You can almost always give your original command, the command "start_point=xxx" and "mr_rosetta_solutions=my_pickle_file.pkl" and it should then continue on from there. I will put this in the docs... All the best, Tom T >> Dear phenixbb, >> >> The Phenix manual mentions several times that making a file named >> STOPWIZARD >> in the wizard root directory will stop the wizard in an orderly fashion >> the >> next time it checks for STOPWIZARD. There is however no mention of how to >> properly restart the wizard after having stopped it... How? I seem to >> remember something about it being >>>phenix.appropriate_wizard params_used_in_wizard.eff >> but I'd like be sure that I don't simply start over by doing that. >> >> Also can I then change the parameters before restarting the wizard? Say, >> hypothetically of course, that I've started mr.rosetta and asked for 1000 >> prerefined models and realised that it will take ridiculously long and be >> rather redundant since I specified both the .hhr file from global and >> local >> alignment and the top hit has two chains (resulting in mr.rosetta wanting >> to >> build 4000 prerefined models) but I don't want to throw away the 700 >> models >> already built. Could I do something about that? >> >> Sincerely, >> Morten Grøftehauge, PhD >> Durham University >> _______________________________________________ >> phenixbb mailing list >> phenixbb(a)phenix-online.org >> http://phenix-online.org/mailman/listinfo/phenixbb >>

1 0

STOPWIZARD
by Morten Grøftehauge 17 Oct '11

17 Oct '11

Dear phenixbb, The Phenix manual mentions several times that making a file named STOPWIZARD in the wizard root directory will stop the wizard in an orderly fashion the next time it checks for STOPWIZARD. There is however no mention of how to properly restart the wizard after having stopped it... How? I seem to remember something about it being >phenix.appropriate_wizard params_used_in_wizard.eff but I'd like be sure that I don't simply start over by doing that. Also can I then change the parameters before restarting the wizard? Say, hypothetically of course, that I've started mr.rosetta and asked for 1000 prerefined models and realised that it will take ridiculously long and be rather redundant since I specified both the .hhr file from global and local alignment and the top hit has two chains (resulting in mr.rosetta wanting to build 4000 prerefined models) but I don't want to throw away the 700 models already built. Could I do something about that? Sincerely, Morten Grøftehauge, PhD Durham University

1 0

mr_rosetta memory usage
by Tjaard Pijning 17 Oct '11

17 Oct '11

Hi, I am currently running mr_rosetta and found that after a while the computer (running OpenSuse/linux) slows down dramatically because its memory (8 GB) is running full. I’m not sure if the program still runs or not. Is there a way to limit the amount of memory that phenix/mr_rosetta uses ? Or should I e.g. lower the amount of models being used (currently 5) ? (Before starting mr_rosetta I ran the regression test which was fine). Kind regards, Tjaard Pijning University of Groningen This is the parameter file : mr_rosetta { input_files { seq_file = gEP_mature.txt" hhr_files = "051011_3011840.hhr" alignment_files = None model_info_file = "" data = "XDS_ASCII_C2221.mtz" labin = None search_models = None copies_in_search_models = None mr_rosetta_solutions = "" ids_to_load = None map_coeffs = "" labin_map_coeffs = None map = "" display_solutions = False fragment_files = "aat000_03_05.200_v1_3" fragment_files = "aat000_09_05.200_v1_3" use_dummy_fragment_files = False sort_fragment_files = True } output_files { log = "mr_rosetta.log" params_out = "mr_rosetta_params.eff" } directories { temp_dir = "mr_rosetta/MR_ROSETTA_7/WORK_1" workdir = "mr_rosetta/MR_ROSETTA_7" output_dir = "mr_rosetta" top_output_dir = "mr_rosetta/MR_ROSETTA_7" rosetta_path = "" rosetta_binary_dir = "rosetta_source/bin" rosetta_binary_name = "mr_protocols.default" rosetta_script_dir = "rosetta_source/src/apps/public/electron_density" rosetta_pilot_script_dir = "rosetta_source/src/apps/pilot/frank/" rosetta_database_dir = "rosetta_database" } read_hhpred { number_of_models = 5 number_of_models_to_skip = 0 copies_to_extract = None only_extract_proper_symmetry = False } place_model { run_place_model = True prerefine { run_prerefine = False number_of_prerefine_models = 1000 number_of_models_in_ensemble = 1 } model_already_placed = False model_already_aligned = False number_of_output_models = 5 align_with_sculptor = True identity = None identity_for_scoring_only = 25 use_all_plausible_sg = True overlap_allowed = 300 selection_criteria_rot_value = 75 fast_search_mode = True search_down_percent = 25 mr_resolution = 3 refine_after_mr = True denmod_after_refine = True find_ncs_after_mr = True min_length_ncs = 10 fixed_ensembles { fixed_ensembleID_list = None fixed_euler_list = 0 0 0 fixed_frac_list = 0 0 0 fixed_frac_list_is_fractional = True } copies_of_search_model_to_place = None } rescore_mr { run_rescore_mr = True nstruct = 5 relax = False include_unrelaxed_in_scoring = False align = True edit_model = False stage_to_rescore = "mr_solution" } rosetta_rebuild { run_rosetta_rebuild = True stage_to_rebuild = "rescored_mr_solution" max_solutions_to_rebuild = 5 min_solutions_to_rebuild = 1 llg_percent_of_max_to_keep = 50 rosetta_models = 100 chunk_size = 1 edit_model = True superpose_model = False } rosetta_rescore { run_rosetta_rescore = True percentage_to_rescore = 20 min_solutions_to_rescore = 2 } similarity { run_similarity = False required_cc = 0.2 number_of_required_cc = 5 } refine_top_models { run_refine_top_models = True stage_to_refine = None sort_score_type = None percent_to_refine = 20 denmod_after_refine = True } average_density_top_models { run_average_density_top_models = True percent_to_average = 100 } relax_top_models { run_relax_top_models = True stage_to_relax = None number_to_relax = 2 nstruct = 5 } autobuild_top_models { run_autobuild_top_models = True number_to_autobuild = 2 quick = False phase_and_build = False macro_cycles = None morph = False edit_model = True use_map_coeffs = True } setup_repeat_mr_rosetta { run_setup_repeat_mr_rosetta = True repeats = 1 template_repeats = 0 morph_repeats = 0 number_to_repeat = 1 acceptable_r = 0.25 minimum_delta_r = None } repeat_mr_rosetta { run_repeat_mr_rosetta = True copies_in_new_search_group = 1 update_map_coeffs_with_autobuild = True } rosetta_modeling { map_resolution = 3 map_grid_spacing = 1.5 map_weight = 1 map_window = 5 include_solvation_energy = True weights_file = "" } crystal_info { resolution = 0 space_group = None chain_type = *PROTEIN DNA RNA ncs_copies = 2 } control { verbose = False debug = False raise_sorry = False dry_run = False nproc = 5 group_run_command = "sh " condor = None single_run_command = "sh " background = None ignore_errors_in_subprocess = True check_run_command = False max_wait_time = 100 wait_between_submit_time = 1 n_dir_max = 100000 number_to_print = 5 write_run_directory_to_file = None resolve_command_list = None start_point = *place_model rescore_mr rosetta_rebuild rosetta_rescore \ similarity refine_top_models average_density_top_models \ relax_top_models autobuild_top_models \ setup_repeat_mr_rosetta repeat_mr_rosetta stop_point = place_model rescore_mr rosetta_rebuild rosetta_rescore \ similarity refine_top_models average_density_top_models \ relax_top_models autobuild_top_models \ setup_repeat_mr_rosetta repeat_mr_rosetta } non_user_params { file_base = None print_citations = True highest_id = 0 is_sub_process = False dummy_autobuild = False dummy_refinement = False dummy_rosetta = False skip_clash_guard = True correct_special_position_tolerance = None comparison_mtz = "" labin_comparison_mtz = None write_local_files = False rosetta_fixed_seed = None } }

3 2

re-building chains
by Leonid Sazanov 14 Oct '11

14 Oct '11

Hi, how exactly can one tell Autobuild to rebuild-in-place chains A and B from the input model, but build chain C de-novo? If I give it list of chains to re-build (rebuild_near_chain="A B" in GUI, dev-874 version), what happens is that either all chains are rebuild-in-place (if rebuild-in-place=True) or all chains are built de-novo (if rebuild-in-place=False). Thanks! -- Dr. Leonid A. Sazanov Research group leader Medical Research Council Mitochondrial Biology Unit Wellcome Trust / MRC Building Hills Road Cambridge CB2 0XY WEB: www.mrc-mbu.cam.ac.uk Tel: +44-1223-252910 Fax: +44-1223-252915

2 1