June 2008 - phenixbb - phenix-online.org

xtriage crash - what to do?
by Martin Moche 09 Jun '08

09 Jun '08

Dear Phenix developers, I got a crash when running phenix xtriage on a mtz-file in P3112. Error message below: The logfile prior to the crash is attached. Best regards, Martin ############# Error message ################ Traceback (most recent call last): File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/command_line/xtriage.py", line 7, in <module> xtriage.run(args=sys.argv[1:]) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/xtriage.py", line 725, in run plot_out = string_buffer ) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/xtriage.py", line 383, in __init__ additional_parameters=self.params.scaling.input.parameters.misc_twin_parameters) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/twin_analyses.py", line 2380, in __init__ out=out, verbose=verbose) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/twin_analyses.py", line 425, in __init__ self.show(out) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/twin_analyses.py", line 596, in show self.suggest_new_space_groups(out=out) File "/net/fileserver/xtal/phenix-1.3b-rc4/mmtbx/mmtbx/scaling/twin_analyses.py", line 450, in suggest_new_space_groups tmp_space_group = sgtbx.space_group_info( str(tmp_space_group), space_group_t_den=t_den) File "/net/fileserver/xtal/phenix-1.3b-rc4/cctbx/cctbx/sgtbx/__init__.py", line 61, in __init__ space_group_symbols=symbols, t_den=space_group_t_den) RuntimeError: cctbx Internal Error: /net/longnose/scratch1/phenix/phenix-1.3b-rc4/cctbx/sgtbx/change_of_basis_op.cpp(70): CCTBX_ASSERT(c_.t().den() % s.t().den() == 0) failure. __________________________________________________ Martin Moche, Ph.D., M.Sc. Senior scientist Karolinska Institutet MBB/SGC Scheeles väg 2 171 77 Stockholm Sweden Phone: +46-8-524 868 43 Mob: +46-733-229327 Fax: +46-8-524 868 68 Martin.Moche(a)ki.se

2 1

disulfides and .defs, maybe custom geom.
by Bryan W. Lepore 09 Jun '08

09 Jun '08

i can start phenix with new .pdb/.mtz, and custom geometry with satisfactory output. problem is, when i provide that output .def file (possibly w/ custom geometry in it) on the command line, phenix always sets up disulfides when it didn't before: Number of disulfides: simple=4, symmetry=0 Simple disulfide: " SG CYS A 14 " - " SG CYS A 14 " distance=0.10 Simple disulfide: " SG CYS A 140 " - " SG CYS A 140 " distance=0.08 Simple disulfide: " SG CYS A 144 " - " SG CYS A 144 " distance=0.54 Simple disulfide: " SG CYS A 147 " - " SG CYS A 147 " distance=0.18 Custom bonds: Sorry: More than one atom selected: "chain A and resname CYS and resid 140 and name SG" Number of selected atoms: 2 my ligands are attached to the cysteines, so i suspect some trouble here but can't see it - shouldn't be a problem, right? if i set disulfide_distance_cutoff= 1.0 or less, phenix drives the system into swap. any pointers, appreciated. -bryan

2 1

effect of solvent fraction on density modification/solvent flattening for a membrane protein
by Pascal EGEA 08 Jun '08

08 Jun '08

Dear All, This is a question about the effect of solvent fraction estimation on the quality of the initial phases. I am trying to solve the structure of a membrane protein at moderate resolution 3.5A from a MAD data set (peak, inflection and high energies). I have one copy of my complex which gives me a solvent fraction of 0.7 without considering the contribution of the detergent. With a related molecule I found a Phaser-MR solution but it does not refine although it is right. I am using Phenix with the phases obtained from molecular replacement and combining them with the MAD signal to locate the seleniums. A visual inspection of the anomalous fourier difference map shows 6 sites out of the 7 expected at 4sigma contouring. If I let Phenix decide by itself it goes on its own with two copies in the ASU and considers a solvent fraction of 0.50; then it finds 7 sites (and not 14 ? why is that?) (the seventh is weak but we expect this for this region of the complex). now if I force Phenix to take ncs_copies=1 and solvent fraction=0.7 then it only find 6 sites. Wether it is 2 "freely chosen" molecules/ASU and 0.5 of solvent or 1 "forced" molecule/ASU and 0.7 of solvent, the 6 first sites are identical (although occupancies are quite higher in the first case) Should I take into account the detergent and lower somewhat my solvent fraction (in the 0.6 range maybe) with the fixed one copy ASU estimation or just trust Solve/Resolve when it runs the density modification part. Does it really affect the quality of the phasing? Between these two extremes I am surprised to see that the quality of the resolve map are quite similar apparently showing clear new structural features. Thanks in advance for your comments, Pascal F. Egea, PhD Post Doctoral Researcher University of California San Francisco Department of Biophysics and Biochemistry Robert Stroud Laboratory pascal(a)msg.ucsf.edu

3 2

OS X source install relocatable?
by Ben Eisenbraun 05 Jun '08

05 Jun '08

Howdy, I'm doing an install from source on Mac OS X from the latest Phenix source packages. We install everything on an NFS mount so we can share it out to the workstations, so it's a little different than most OS X installs I suspect. Anyway, the installer was failing like this: (from the top level source directory) ./install --prefix=/programs/i386-mac/phenix/1.3-rc2 Which was getting me: configuring python compiling python csh: /programs/source/phenix/1.3-rc2/osx86/phenix-installer-1.3/bin/check_lib: No match. installation of python failed problem building main python library Looking at bin/install-from-source, it was failing on line 482: set py_lib = `$PHENIX_TAR/bin/check_lib $PHENIX_BUILD/base/Python.Framework/Versions/Current/lib/python*/config python` The reason for the failure is that the python framework installs with a lower case 'f': /programs/i386-mac/phenix/1.3-rc2/phenix-1.3-rc2/build/mac-intel-osx/base/Python.framework/ On a default HFS+ volume, which is case insensitive, this works just fine, but it fails when you try to install on a case-sensitive volume. Editing bin/install-from-source and :%s/Python.Framework/Python.framework/g seems to have fixed the build problem. -ben -- Ben Eisenbraun Structural Biology Grid Harvard Medical School http://sbgrid.org http://hms.harvard.edu

1 0

constraining alpha-helical geometry in phenix
by Pascal EGEA 05 Jun '08

05 Jun '08

Dear All, I am currently refining a structure at moderate resolution (~3.4A) which is mainly composed of alpha-helices . Is there a way to constrain the backbone geometry of the alpha helical regions in the model used for refinement. Many Thanks, Pascal F. Egea, PhD Post Doctoral Researcher University of California San Francisco Department of Biophysics and Biochemistry Robert Stroud Laboratory pascal(a)msg.ucsf.edu

2 1

System Administrator Position in Berkeley California
by Paul Adams 04 Jun '08

04 Jun '08

System Administrator 3 Position =============================== The Joint BioEnergy Institute is seeking a Systems Administrator, to be a part of the Technology Division. The scope of this position includes managing scientific computing resources, primarily linux- based clusters and servers, making project and public data accessible in a timely manner to project participants, and integrating the data with existing data from other databases. The Joint BioEnergy Institute (JBEI) is a new multi-organizational research center comprising world-class resources and expertise located in one facility at Emeryville, California. The goal of JBEI is to use rapidly advancing scientific techniques, such as systems and synthetic biology, to accelerate development of the nation's biofuels industry. JBEI is a U.S. Department of Energy funded Bioenergy Research Center, managed by Lawrence Berkeley National Laboratory (LBNL) with partners from 5 internationally recognized scientific institutions: Sandia National Laboratories (SNL), Lawrence Livermore National Laboratory (LLNL), University of California-Davis (UCD), University of California- Berkeley (UCB), and the Carnegie Institution for Science Duties: Plan, organize, prioritize and complete assigned tasks and projects in a timely manner. Frequently and clearly communicate task or project status to customers to either set or negotiate expectations. Under the broad supervision of the Vice President of Technology and the Computational Analysis Director: Provide linux system administration and user support for research groups. This includes system installations, software installations from packages or from source code, distribution upgrades, OS and security patches, system monitoring, disk and file system management, system hardware and peripheral management, backup and recovery management, troubleshooting, and crash recovery. Configure, administer, and troubleshoot various compute clusters, web and database servers, and other services and daemons. Participate in the formulation, planning, organizing, implementation and support of new services or capabilities. Provide basic support for linux desktop systems, including hardware and software installations and upgrades. Support a group of approximately 10 full-time programmers and computational biologists in using JBEI computational resources. Key Success Factors: At least five years linux system administration experience, with responsibility for a wide range of hardware and distributions. Specific areas of required experience include compute cluster and job queuing software administration and support; Apache; Python/Perl; bash/ tcsh, including writing complex Bourne shell scripts; runlevel 1 for critical update or recovery purposes; remote administration, including ssh, remote serial consoles, and remote PDUs; NFS; and monitoring various models of RAID controller. Firm understanding of linux internals including the boot process, kernel versions, and recovery scenarios. Experience with a code management system, preferably CVS or Subversion, as both administrator and code contributor. Experience with relational database operation and maintenance, specifically with MySQL, PostgreSQL or Oracle. Database experience should include database design, implementation, installation, and administration, including backup and restore of large databases. Demonstrated ability to make progress on multiple tasks simultaneously, and to work both independently and as a member of a team. Must be available as needed nights and weekends for critical system updates that cannot be performed during regular lab hours. Application Procedure: This is a full-time one-year Term appointment, renewable under the same conditions, contingent upon the continued availability of funding and work. For consideration apply online at: http://jobs.lbl.gov, select "Search Jobs", and enter job number 21565 in the keyword search field. For more information about JBEI go to: www.jbei.org -- Paul Adams Deputy Division Director, Physical Biosciences Division, Lawrence Berkeley Lab Adjunct Professor, Department of Bioengineering, U.C. Berkeley Vice President for Technology, the Joint BioEnergy Institute Head, Berkeley Center for Structural Biology Building 64, Room 248 Tel: 1-510-486-4225, Fax: 1-510-486-5909 http://cci.lbl.gov/paul Lawrence Berkeley Laboratory 1 Cyclotron Road BLDG 64R0121 Berkeley, CA 94720, USA. Administrator: Delia Clark [ DNClark(a)lbl.gov ][ 1-510-486-6331 ] --

1 0

optional bits of the CCI Apps installion?
by Ben Eisenbraun 03 Jun '08

03 Jun '08

Hi Phenix Developers, When I install CCI Apps, I end up with a directory that looks like this: $ ls -1 1NSF.pdb cci_apps_build cci_apps_fedora3_py251.selfx cci_apps_install_script.csh cci_apps_sources nsf_d2_peak.sca test_apps Which of the bits are optional and can be deleted without affecting the installation? It seems like the selfx install file is fine to delete. How about the sources directory? $ du -k --max-depth=1 62584 ./cci_apps_sources 125840 ./cci_apps_build 225012 . If the sources directory is not needed, that would save me about 30% of the disk use for the installation, and since I have 2-3 concurrent versions installed, and those versions get pushed around to ~75 machines across the network, every bit I can trim helps. As a side note, I think it might be clearer to refer to the "Opteron" builds of the CCI Apps as AMD64 or x86_64. I didn't see anything on the download page that makes this explicit. Thanks. -ben -- Ben Eisenbraun Structural Biology Grid Harvard Medical School http://sbgrid.org http://hms.harvard.edu

2 1

Group ADP not working
by alebus＠pasteur.edu.uy 03 Jun '08

03 Jun '08

Dear all, I am refining a structure (2.9angs res; P21) with phenix (Version: 1.3b; Release tag: rc6; linux platform) I want to carry on group_ADP refinement, according to the ('lengthy') command line : phenix.refine eden_unique1.hkl hlcoeff_blur.hkl abril28_phenix.pdb input.experimental_phases.labels="PA_BLUR PB_BLUR PC_BLUR PD_BLUR" ordered_solvent=false simulated_annealing=false simulated_annealing.mode=first strategy=rigid_body+individual_sites+group_adp+tls main.ncs=tru e main.number_of_macro_cycles=10 xray_data.low_resolution=15.0 xray_data.high_resolution=2.7 tls_group_selections.params rigid_body_selections.params group_anomalous_1.params elbow.acp_A_pdb.001.cif restraints_edits.params convert_to_isotropic=true everything works just fine (including correct handling of an 'unusual' ligand; selected NCS restraints; TLS; etc)...except for the group ADP : phenix keeps doing individual atom B-factor refinement (e.g. these are some pertinent remarks in the output pdb REMARK ****************** REFINEMENT STATISTICS STEP BY STEP ****************** REMARK leading digit, like 1_, means number of macro-cycle REMARK 0 : statistics at the very beginning when nothing is done yet REMARK 1_bss: bulk solvent correction and/or (anisotropic) scaling REMARK 1_xyz: refinement of coordinates REMARK 1_rbr: rigid body refinement REMARK 1_gbr: group B-factor refinement REMARK ------------------------------------------------------------------------ REMARK R-factors, x-ray target values and norm of gradient of x-ray target REMARK stage r-work r-free xray_target_w xray_target_t REMARK 0 : 0.3951 0.4346 -6.345353e+00 -6.295977e+00 REMARK 1_bss: 0.3189 0.3519 -6.182904e+00 -6.132416e+00 REMARK 1_rbr: 0.3199 0.3553 -6.164872e+00 -6.111524e+00 REMARK 1_bss: 0.3197 0.3548 -6.158236e+00 -6.106543e+00 REMARK 1_xyz: 0.3145 0.3612 -6.243414e+00 -6.163029e+00 REMARK 1_adp: 0.2890 0.3327 -6.111496e+00 -6.011406e+00 REMARK 2_bss: 0.2862 0.3290 -6.107396e+00 -6.004629e+00 REMARK 2_xyz: 0.2801 0.3278 -6.104559e+00 -5.995221e+00 REMARK 2_adp: 0.2783 0.3213 -6.062947e+00 -5.954846e+00 REMARK 3_bss: 0.2783 0.3214 -6.063301e+00 -5.955103e+00 REMARK 3_xyz: 0.2755 0.3218 -6.065099e+00 -5.950739e+00 REMARK 3_adp: 0.2751 0.3218 -6.062715e+00 -5.947567e+00 ... and so on, always doing 'X_adp' (and not 'X_gbr') on the different steps... starting from a single value on the B-factor column in the input pdb, of course I end up having individually varied figures in the end, which I don't want I will greatly appreciate any help on what am I doing wrong. Thanks, Alejandro

2 1

phenix.elbow
by Morten Grøftehauge 02 Jun '08

02 Jun '08

Dear Phenix.BB, I wrote a mail earlier about some problems I had with phenix.elbow but it didn't seem to get through. I've solved two of the problems now but also found a new one so I'll try posting again. When you set the --xyz option along with --clean then your .xyz file is deleted. Of course the easy solution is simply to not use --clean but I imagine that it is a minor bug since it doesn't happen with --tripos. Secondly I have a slightly larger problem. When an isoSMILES with more than one molecule is inputted then phenix.elbow joins them up to a single molecule (or places them so closely that all programs, including phenix.elbow, interprets them as one molecule when converting to a format such as mol2 with explicit bonds). For an example try phenix.elbow --output=009_3085017 --write_hydrogens=false --overwrite --clean --tripos --smiles='C=CCN.C1C(O1)CCl' The period is supposed to designate two seperate molecules. Remember to use elbow.become_expert before running that line. I hope someone can help me. Sincerely, -- Morten K Grøftehauge PhD student Department of Molecular Biology Gustav Wieds Vej 10 C 8000 Aarhus C - Denmark Phone: +45 89 42 52 61 Fax: +45 86 12 31 78 www.bioxray.dk

3 2

shell variable SSH_Path not set
by shivesh kumar 02 Jun '08

02 Jun '08

Dear all, We have installed Phenix in Red Hat Linux enterprises 5. It takes a long time for the GUI to appear with the following message-- Shell variable SSH_Path not set What is the solution of this problem. Thanx in advance for suggestions. shivesh

2 1