- phenixbb - phenix-online.org

phenix.refine 1.3 vs. 1.4 map quality ?
by William G. Scott 24 Aug '09

24 Aug '09

Hi folks: About a year ago I refined a 1.6 Å RNA structure with phenix.refine 1.3b-rc6 and then got distracted. I picked it up today and essentially repeated the last round of refinement with 1.4-57 (and also 1.4-159). The maps are subtly different, but consistently slightly worse with 1.4. Here are two examples: http://sage.ucsc.edu/~wgscott/mystuff/old_vs_new.pdf In the second example, this would lead to deletion of one of the octahedrally coordinated waters on a known Mg++ ion. 2OEU is similar, except 2.0 Å resolution, and it has Mn2+ instead of Mg ++. A similar thing happens with the corresponding water in that structure. Has the weighting for the bulk solvent mask or something like that been increased? Thanks. Bill Scott

5 8

difference in B factor for NCS groups
by James Whittle 21 Aug '09

21 Aug '09

Dear all, I'm working with a structure were one NCS copy has significantly high B-factors than the other. That one copy makes few crystal contacts to stabilize it. Is there a recommended way to adjust the refinement to reflect this ... using TLS? Decreasing b_factor_weight? Is there a way refine a group_b to be added to the b-factor of each atom in that chain? --James

3 2

Error in running Autosol....
by S. Karthikeyan 20 Aug '09

20 Aug '09

Dear All, I am using phenix-1.4-138 in RHEL5 Xeon-64 bit workstation. I am trying to solve a structure of Se-Met through phenix.autosol_gui. Once the programs starts it goes to several stages including converting from mtz to sca, xtriage and so on. Finally it gives an error message as follows: ******************************************************************************** Failed to carry out AutoSol_scale_and_analyze_mad: None of the solve versions worked See overall log file and also solve log file (in AutoSol_run_2_/LAST.LOG) ******************************************************************************** I found the LAST.LOG file is empty. I am trying to find out what could be problem? Any suggestions or help is much appreciated! Thanking you S.Karthikeyan

1 1

Problem with LLP ligand
by James Naismith 19 Aug '09

19 Aug '09

Dear All, I am trying to refine a structure which has a Lys linked to PLP. In the PDB this is called LLP. When I take it through refine it crashes complaining about the ligand. "Sorry : Fatal problems interpreting PDB file: Number of atoms with unknown nonbonded energy type symbols: 8 Please edit the PDB file to resolve the problems and/or supply a CIF file with matching restraint definitions, along with apply_cif_modification and apply_cif_link parameter definitions if necessary (see phenix.refine documentation). Also note that phenix.elbow is available to create restraint definitions for unknown ligands. Traceback: None " I know I have to use REEL to fix this in some way. I have used REEL to make ligands not bound to protein before. It is how to do this when the ligand is covalently bound that I am stuck with. Any help (particularly a cif restraints file!) gratefully received. best Jim ------------------------------------------------------------------------------- Major third-party components of Phenix include: Python, wxwidgets, wxPython, Boost, SCons, Clipper, CCP4 Monomer Library, CCP4 I/O libraries, PyCifRW, FFTPACK, L-BFGS Enter phenix.acknowledgments for details. ------------------------------------------------------------------------------- Processing inputs. This may take a minute or two. Working crystal symmetry after inspecting all inputs: Unit cell: (50.677, 137.283, 62.258, 90, 109.637, 90) Space group: P 1 21 1 (No. 4) Monomer Library directory: "/usr/local/phenix-1.4-115/chem_data/mon_lib" Total number of atoms: 13558 Number of models: 1 Model: "" Number of chains: 3 Chain: "A" Number of atoms: 6752 Number of conformers: 2 Conformer: "B" Number of residues, atoms: 420, 6747 Unexpected atoms: {'LLP,HD3': 1, 'LLP,HG3': 1, 'LLP,HB3': 1, 'LLP,HE3': 1} Classifications: {'peptide': 420} Link IDs: {'PTRANS': 14, 'TRANS': 401, 'PCIS': 4} Chain breaks: 3 Conformer: "A" Number of residues, atoms: 420, 6747 Unexpected atoms: {'LLP,HD3': 1, 'LLP,HG3': 1, 'LLP,HB3': 1, 'LLP,HE3': 1} Classifications: {'peptide': 420} Link IDs: {'PTRANS': 14, 'TRANS': 401, 'PCIS': 4} Chain breaks: 3 bond proxies already assigned to first conformer: 6820 Chain: "B" Number of atoms: 6800 Number of conformers: 1 Conformer: "" Number of residues, atoms: 423, 6800 Unexpected atoms: {'LLP,HD3': 1, 'LLP,HG3': 1, 'LLP,HB3': 1, 'LLP,HE3': 1} Classifications: {'peptide': 423} Link IDs: {'PTRANS': 14, 'TRANS': 404, 'PCIS': 4} Chain breaks: 3 Chain: "M" Number of atoms: 6 Number of conformers: 1 Conformer: "" Number of residues, atoms: 6, 6 Unusual residues: {' SM': 6} Classifications: {'undetermined': 6} Link IDs: {None: 5} Number of atoms with unknown nonbonded energy type symbols: 8 "ATOM 3149 HG3 LLP A 211 .*. H " "ATOM 3151 HE3 LLP A 211 .*. H " "ATOM 3153 HD3 LLP A 211 .*. H " "ATOM 3155 HB3 LLP A 211 .*. H " "ATOM 9896 HG3 LLP B 211 .*. H " "ATOM 9898 HE3 LLP B 211 .*. H " "ATOM 9900 HD3 LLP B 211 .*. H " "ATOM 9902 HB3 LLP B 211 .*. H " Time building chain proxies: 9.24, per 1000 atoms: 0.68 The University of St Andrews is a charity registered in Scotland : No SC013532

3 2

RESOLVE with RNA/protein structure at low resolution
by Peter Grey 19 Aug '09

19 Aug '09

Hello all, I try to solve the structure of RNA and RNA/protein complex at low resolution (5A). Phases come from partial model MR solution. I would really welcome advice about density modification in such cases with RESOLVE.. Does the RESOLVE database of density distributions contain RNA/protein examples ? Can RESOLVE work at low resolution given that histograms profiles are at resolutions from 1.2 A to 4 A ? Many thanks for you help, Peter.

2 1

AutoBuild help
by Timothy C. Umland 18 Aug '09

18 Aug '09

I am new to running the AutoMR and AutoBuild Wizards, and I am having a problem getting AutoBuild to run. I have run AutoMR and it appeared to run fine, yielding a solution. However, when it did the default hand-off of the MR solution to AutoBuild, AutoBuild crashed. I then tried to run AutoBuild by itself, using the MR solution as the input model. Again, AutoBuild crashed. The error was the same in both cases. I was running AutoMR and AutoBuild from the command line. I am running Phenix version 1.4-3 on an Intel Mac (OS X 10.5.8). The AutoBuild error was **************************************************************************** **** Failed to carry out AutoBuild_build_cycle: child process stderr output: command: 'csh /AutoBuild_run_3_/TEMP0/RUN_FILE_1' sort: invalid option -- 4 Try `sort --help' for more information. **************************************************************************** **** I tried the suggested "sort --help", but at least to me it wasn't helpful. I am guessing that I have some incompatibility with my local installation of "sort" and Phenix? Any help is appreciated. Thanks, Tim

2 2

1-2 October: What is a Macromolecular Complex ? meeting at the NKI, Amsterdam
by Anastassis Perrakis 18 Aug '09

18 Aug '09

Dear all, This is a reminder and some news for the meeting: "What is a macromolecular Complex? Shades of Meaning Across Cellular, Systems, and Structural Biology" http://xtal.nki.nl/Oct2009 - Registration closes on Monday 7th September at 12:00 noon. No exceptions after the deadline. - There will be eight (8) fellowships for attending the meeting, 500 Euro (max) each (*) Best regards on behalf of the organizers, Tassos (*) Only available if your lab is NOT part of the EU programs that support the meeting (SPINE-2-Complexes, TEACH-SG, 3D-Repertoire) and you really wanted to attend but could not for financial reasons, but 500 Euro will make it possible for you to join us, please email a.perrakis.AT.nki.nl with a short motivation and research interests, 10-15 lines max (no attachments please, simple text). Preference will be given to people not yet registered, since we might well assume that people that already registered had the money already sorted. The amount is the MAXimum available per person for ALL expenses, and it will be reimbursed after the meeting, *** based on receipts ***!

1 0

Re: [phenixbb] Error in running Autosol....
by Thomas C. Terwilliger 15 Aug '09

15 Aug '09

Hi Karthikeyan, I'm sorry for the trouble! I see that you are running the very latest phenix from last night. This version passed our automated tests and it appears fine in general. Somehow the solve scaling step is not working in your run. I haven't seen this happen before, but my guesses are: 1. I have been making small changes in the autosol printout and it is possible that there is a place where solve is printing something out and the format does not match the variable type, and I didn't catch it in our tests. If this is the case you should be able to find it by running from the command-line and adding the keyword debug=true and it should print out a clearer error message for you that you can send me. You can run from the command line in the same place you ran from the GUI by saying phenix.autosol autosol.eff debug=True because the GUI writes out autosol.eff when it runs, saving your inputs. Then you can override those values by putting debug=True on the command line. Other less-likely possibilities: 2. your input datafiles are of a type that was unexpected (normally .sca, .mtz are fine, but perhaps the columns in an mtz are unusual). 3. Your structure is very large or high-resolution. 4. Something is unusual in your setup so that not much memory is available. If none of these helps, you are welcome to send me the files and I'll check it out. All the best, Tom T >> Dear All, >> >> I am using phenix-1.4-138 in RHEL5 Xeon-64 bit workstation. >> >> I am trying to solve a structure of Se-Met through phenix.autosol_gui. >> Once the >> programs starts it goes to several stages including converting from mtz to >> sca, >> xtriage and so on. >> Finally it gives an error message as follows: >> >> ******************************************************************************** >> Failed to carry out AutoSol_scale_and_analyze_mad: >> >> None of the solve versions worked >> See overall log file and also solve log file (in AutoSol_run_2_/LAST.LOG) >> ******************************************************************************** >> I found the LAST.LOG file is empty. >> >> I am trying to find out what could be problem? Any suggestions or help is >> much >> appreciated! >> >> Thanking you >> >> S.Karthikeyan >> >> >> >> >> _______________________________________________ >> phenixbb mailing list >> phenixbb(a)phenix-online.org >> http://www.phenix-online.org/mailman/listinfo/phenixbb >>

2 1

Re: [phenixbb] low resolution data/parameter ratio
by Pavel Afonine 14 Aug '09

14 Aug '09

Hi Gerwald, > Two more quick questions: > how would you identify the best strategy? I would look at Rfree and Rfree-Rwork, and maps of course. To help in deciding what are the "good / reasonable" values, use POLYGON (Acta Cryst. D65, 297-300 (2009)) (phenix.polygon), local real-space correlation (phenix.real_space_correlation). I think Nat bundled them all into his magic tool called phenix.validation with will do this and more (it is GUI based). > I would aim for the lowest number of parameters with reasonably low > Rfree? Yes, this sounds right in general. > Does phenix report the total number of parameters refined anywhere in > the log? It's trivial and not at the same time (at least the way I understand it, of course I might be very wrong). It's trivial to compute the number of refinable parameters given the strategy you choose, but how you will count the restraints which are kind of quasi-observables, and what about weight that if chosen too strong so it can turn the restraints into quasi-constraints? Pavel.

2 2

low resolution data/parameter ratio
by gerwald jogl 14 Aug '09

14 Aug '09

Hi All, I am struggling with low resolution data (3.6 A) and I am concerned about overfitting. I have a ratio of about 3 reflections per atom. Possible refinement scenarios would be a) coordinates+tls, b) coordinates, group_adp, tls or c) coordinates, individual_adp, tls. The last scenario seems to be pushing the limits. I wonder if there are studies, thoughts or opinions out there on which refinement scenario to use. Thanks for any comment on this, Gerwald

2 1