- phenixbb - phenix-online.org

reference model restrain error
by Stephan Barden 23 Nov '11

23 Nov '11

Dear all, I'm facing a problem with a fresh (yesterday) installation of phenix 1.7.2-869 refining a structure with 6 molecules per ASU at 3.25 A and using a different reference structure with 1 mol/ASU at higher resolution. I wrote an additional parameter file to assign the reference model to each of the six molecules, which worked well with an older version of phenix. After the upgrade neither taking the additional parameter file, nor auto alignment worked for me. The error comming up is always: RuntimeError : Duplicate sequence number and insertion code. Traceback: File "/sw/phenix-1.7.2-869/cctbx_project/libtbx/thread_utils.py", line 229, in run return_value = self._target(self._args, self._kwargs, self._c) File "/sw/phenix-1.7.2-869/cctbx_project/libtbx/runtime_utils.py", line 72, in __call__ result = self.target() File "/sw/phenix-1.7.2-869/phenix/phenix/refinement/runtime.py", line 67, in __call__ call_back_handler=callback) File "/sw/phenix-1.7.2-869/phenix/phenix/refinement/command_line.py", line 75, in run overwrite = command_line_interpreter.command_line.options.overwrite) File "/sw/phenix-1.7.2-869/phenix/phenix/refinement/driver.py", line 209, in __init__ log=self.log) File "/sw/phenix-1.7.2-869/cctbx_project/mmtbx/torsion_restraints/reference_model.py", line 145, in __init__ self.get_reference_dihedral_proxies() File "/sw/phenix-1.7.2-869/cctbx_project/mmtbx/torsion_restraints/reference_model.py", line 611, in get_reference_dihedral_proxies log=self.log) File "/sw/phenix-1.7.2-869/cctbx_project/mmtbx/torsion_restraints/reference_model.py", line 398, in process_reference_groups raise e Has anyone observed a similar problem? Are there any suggestions? Thanks, StephanB.

2 1

solve/resolve problem
by dingding830106＠sina.com 22 Nov '11

22 Nov '11

Dear all when I use SOLVE to deal with the SAD data, it will stop with a error message: “STOP at /net/longnose/scratch2/phenix/phenix-1.7.2-869/solve_resolve/solve/solve_2.cpp(3133)” Can you tell me what happen? wding

2 1

How to reduce clashscore value
by CelinaSocek 22 Nov '11

22 Nov '11

Hi all, I'm now using phenix to refine a small protein about 300 residues long. The resolution of my data is about 2.9 angstrom. Currently I got a refined model with good R-factors about 23/26, RMSD very reasonable and no Ramanchandran outliers. However, I found that the clashscore is extremely high: 63!! And rotamer outliers are also a lot: 8.5%. I wonder, what kind of stretagy I can use to reduce this? Thank you very much! best, Celina

6 16

test_hhr
by Alexander Batyuk 21 Nov '11

21 Nov '11

Dear all, The phenix_regression.wizards.test_command_line_rosetta_quick_tests command hangs at the test_hhr step in phenix-dev-910 on Mac OS X 10.6.8. What could be the reason for that? What kind of log files should I post? The firewall allowed to fetch files from the net. Best wishes, Alex -- Alex Batyuk The Plueckthun Lab www.bioc.uzh.ch/plueckthun

2 2

Re: [phenixbb] How to reduce clashscore value
by CelinaSocek 20 Nov '11

20 Nov '11

Thank you Nat! I will try to fix as many clashes as I can in coot, and hope that I can manage to reduce it to around 20. ;) Also thank other suggestions and discussions! best, Celina > > Which version are you using? We made some changes over the summer > that improve the geometry, especially the clashscore - these are > present in version 1.7.2. If you're still using an earlier version, > this might largely fix the problem. However, if there is > conformational strain in the model, it is difficult to fix severe > clashes without further rebuilding. You should inspect the clashes in > Coot (the Phenix GUI has shortcuts for this, but Coot alone also has > this capability) and fix as many as possible yourself. It is > recommended that you use explicit hydrogen atoms when doing this, > since the reason for clashes is often not obvious if only heavy atoms > are used. (You don't necessarily need to refine with heavy atoms > though - they're just an aid to visualizing clashes.) > > The rotamer problem may be related; there isn't much you can do with > these except fix them manually, but relieving some of the clashes may > help. The dihedral angle restraints simply aren't that strong, so in > areas of weak density it can be difficult to completely prevent > outliers. > > -Nat > > > ------------------------------ > >

1 0

multiple ligand conformations
by Leonard Thomas 20 Nov '11

20 Nov '11

Hello All, One of the groups here has run into a problem in the refinement of their structure or more specifically a ligand bound to the structure. Through a series of other experiments they know that a photo reaction occurs to the compound when exposed to high flux as found at your everyday synchrotron source. The exposure changes the bonding such that there are two conformations with a bit in between obviously. They collected 360 degrees worth of data and if the data is split you can model the two conformations depending which half of the data you use. They are trying to refine with all of the data and modeling of the ligand into the two conformations. In refmac they can model this with no problem but in Phenix the ligand adopts a chemically wrong conformation, sort of a halfway point between the two chemically relevant conformations. They have used Phenix to do similar refinements on other structures but it appears either the last release or the latest release is causing problems. They have started out with everything from 50% occupancy to shifting it towards on one conformation or the other and the ligand always ends up in the chemically wrong position. Data is in the 1.4-1.5 ang. range. Ideas? Thanks in advance. Len Leonard Thomas Ph.D. Macromolecular Crystallography Laboratory Manager University of Oklahoma Department of Chemistry and Biochemistry Stephenson Life Sciences Research Center 101 Stephenson Parkway Norman, OK 73019-5251 lmthomas(a)ou.edu http://barlywine.chem.ou.edu Office: (405)325-1126 Lab: (405)325-7571

2 1

multiple ligand conformations
by Yuri 20 Nov '11

20 Nov '11

Hi, When you say "wrong postion", what does it mean? are the 2 ligands you are trying to refine being "pushed away" from each other? If that is the case first thing I would check is the ALTLOC comlumn of their pdb file entries. Make sure one is ALTLOC A and the other ALTLOC B. I have refined with ligands in four different positions before using phenix before and got satisfactory results... Hope this helps, -- Yuri Pompeu

1 0

MR_ROSETTA
by Wei Shi 18 Nov '11

18 Nov '11

Hi all, I am trying to run phenix.mr_rosetta using the following script for rebuilding the model with Rosetta before MR from phenix website except omitting specifying ncs_copies number. Following is the script I am using: phenix.mr_rosetta \ seq_file = seq.dat \ data=coords1.mtz \ search_models=coords1.pdb \ run_prerefine=True \ number_of_prerefine_models=1000 \ fragment_files = test3.gz \ fragment_files = test9.gz \ rescore_mr.relax=False \ rosetta_models=20 \ space_group=p222 \ use_all_plausible_sg=False \ nproc=4 \ group_run_command=qsub\ When I run the above script, it gives the following error message: child process stderr output: command: 'qsub /home/wei/MR_ROSETTA_14/GROUP_OF_PLACE_MODEL_1/RUN_FILE_1' /bin/sh: qsub: not found Then, I omit the last line in the command “group_run_command=qsub\”. MR_ROSETTA seems to be running. But, it takes forever to run. My questions is: 1. Does anyone happen to know why it tells me “qsub: not found” when I include “group_run_command=qsub” in my script? 2. When I omit “group_run_command=qsub” in the script, mr_rosetta seems to be running. But it just gives me the following message and it takes forever to run. Almost 10h later, no job is finished. I don’t know whether this is normal. If not, any idea about why it’s so slow and any suggestions about how to speed up the run? Thank you so much! Splitting work into 1 jobs and running with 4 processors using sh background=True in /home/wei/MR_ROSETTA_15/GROUP_OF_PLACE_MODEL_1 Starting job 1...Log will be: /home/wei/MR_ROSETTA_15/GROUP_OF_PLACE_MODEL_1/RUN_FILE_1.log

7 9

error: Duplicate selections
by Sebastiano Pasqualato 16 Nov '11

16 Nov '11

Hi all, sorry guys, this might be a silly one, but I can't see where I'm making the mistake. Whenever I turn on TLS refinement, I get this error message. And, as you can see, the Log file doesn't tell me much more. I have generated the TLS groups wit the 'TLS groups selection' tool in phenix. Any hint? Thanks in advance, ciao s -- Sebastiano Pasqualato, PhD Crystallography Unit IFOM-IEO Campus Cogentech - Consortium for Genomic Technologies via Adamello, 16 20139 - Milano Italy tel +39 02 9437 5172 fax +39 02 9437 5990

2 2

Scattering from riding hydrogens?
by Damian Ekiert 16 Nov '11

16 Nov '11

Hello, I have a question regarding two parameters and how they affect riding hydrogens. contribute_to_f_calc = True high_resolution_limit_to_include_scattering_from_h = 1.6 At lower resolution, setting contribute_to_f_calc = false is frequently helpful when using riding hydrogens and has been discussed before. But what is high_resolution_limit_to_include_scattering_from_h = 1.6 doing? If scattering from hydrogens is only included at higher than 1.6A resolution, it would seem like contribute_to_f_calc = True/False would not have any effect at say, 3.5A, but comparing identical refinement runs with contribute_to_f_calc = True or contribute_to_f_calc = False clearly gives different results. Can hydrogens contribute to Fcalc in a way aside from their scattering, or is their contribution accounted for independent of the position of the riding hydrogen (e.g., additional scattering term added to the associated heavy atom)? I think I am misunderstanding what these two parameters are doing and would appreciate some clarification. Thanks, Damian Ekiert

4 5

Multiple TLS groups per molecule
by Yarrow Madrona 15 Nov '11

15 Nov '11

Hi, I would like run TLS and individual isotropic refinement. I want to separate one molecule into four TLS groups. I havn't found documentation for this. Is the number of TLS groups automatically done by phenix refine when you tell it to run TLS refinement on the whole molecule? Thanks for your help -Yarrow -- Yarrow Madrona Graduate Student Molecular Biology and Biochemistry Dept. University of California, Irvine Natural Sciences I, Rm 2403 Irvine, CA 92697

2 3

Small issue with residue PTR
by Terry Lang 15 Nov '11

15 Nov '11

Hey Everyone, I have found a small issue with atom naming for residue PTR. When phenix.ready_set adds hydrogens to the residue, it names the hydrogens connected to the CB atom to be HB3 and HB2. However, the hydrogens in the cif file located in chem_data/mon_lib are labeled HB1 and HB2. Manually changing either the cif file or the atom name in the pdb file solves the problem. Terry -- P. Therese Lang Post Doc Alber Lab, UC Berkeley

2 1

How can I add another chain to my PDB by Coot
by Dialing Pretty 14 Nov '11

14 Nov '11

Dear All, By Phenix refine I got a PDB (Only one chain) and mtz output files. After analysis by Coot I find I need to add a protein fragment so that a large blob will disappear. I know I use Baton to construct the protein fragment. Will you please tell me how can I input the sequence file of the protein fragment so that I can refine the protein fragment by Coot? I am looking forward to getting your reply. Regards. Dialing

2 1

twin/pseudo-symmetry question
by Green, Todd 12 Nov '11

12 Nov '11

Hello all- I have a question regarding the possibility of twinning and or psuedo-symmetry. I have a crystal that processes well in R32, ie. less than 6% r-merge at 2.2 angstrom. We collected multiple potential derivatives and were able to get some phasing from one that was good enough to model in the helices. The strange thing is that when looking at the harker sections there were a lot of unexplained strong peaks. And there is a very narrow range of parameters that we could get the sites with Shelx and end up with interpretable maps(i know that this part might not be unusual). My first thought was that there could be twinning. Since there is no twin law for this space group, I thought that it might suggest that there was twinning in a lower symmetry space group and/or that we have a pseudo-R32 crystal. Processing works well in C2 and R3, as expected. In those cases Xtriage suggests that the real symmetry is likely to be R32. In the case of C-2, xtriage suggetss if it is twinned then it is about 45% twinned. That's the background. Fast forward with the help of phenix Autobuild and some manual correction, we have been able to refine the structure with the R32 processed data(with tls) to r_work = 0.1894, r_free = 0.2287, bonds = 0.007, angles = 0.908 at 2.2 angstrom with almost everything accounted for. I think I should be quite comfortable with this. BUT, I'm still wondering about those Harker sections. I've been reading the manuscripts of Dauter regarding twinning and I think I am at this point more curious than anything. But not leaving well enough alone, I went back to the data processed in C2. I can easily do molecular replacement, following refinement at 2.3 angstrom(I need a lot more frames for this sg so the cutoffs are a bit lower but r-merge is ~ the same as R32) with TLS I get r_work = 0.1850, r_free = 0.2249, bonds = 0.007, angles = 0.946. this doesn't seem significant to me. However if I refine with the two possible twin operations, the values drop with one being better than the other. both have the same starting model: r_work = 0.1639 r_free = 0.1880 bonds = 0.006 angles = 0.902 r_work = 0.1582 r_free = 0.1866 bonds = 0.006 angles = 0.891 The twin fraction from the refined pdb header is 0.544 for the better statistical model and the r-values seem too low for 2.3 angstrom data. Is the high twin fraction suggestive of the higher symmetry space group or inconclusive towards that. of interest is that the maps look better to me than the r32 maps. by that i mean that the maps don't have what i would call minor dots of negative density for some(not a lot) of the side chains as we see for some side chains in the R32 case. also some side chains look better in the C2 twin refined case. Lastly for some of the monomers(there are now 6 copies versus 2 in the R32 case), I can add an additional residue or two on the termini which are not seen before. I wonder if this also suggests that the true space group is C2. Are these r-values too low? does the difference in values according to each twin law mean something? In the end, the models are essentially the same in either space group (with the exception of a couple of residues at the termini). So, I don't think I learn much one way or the other. But the curiocity keeps coming back. So is it likely to be R32 or C2 or C2 with a hefty twin fraction? Anyone have an explanation? Thanks in advance- Todd

3 7

peptide link problem solved LINKR in pdb is not recognized but LINK is
by Christian Roth 10 Nov '11

10 Nov '11

Hi, In my pdb from Refmac the keyword LINKR is used but only LINK is interpreted by phenix. I changed this and now it works as it should. Sorry, a careful analysis sometimes help. Regards Christian

1 0

peptide link
by Christian Roth 10 Nov '11

10 Nov '11

Hi, I have an modified aminoacid in my peptide chain (backbone modified). I have a cif file and I refined it previously in Refmac. I put the Link definition in the pdb and Refmac has interpreted the peptide link as I wanted. Now I did a run in phenix to compare the results and the chain is broken I created the SNN cif file with ReadySet Reel and I tried to incorporate the link definition via GUI in the following way apply_cif_link: data_link: TRANS residue_selection_1: chain A and resseq 366 residue_selection_2: chain A and resseq 367 apply_cif_link: data_link: TRANS residue_selection_1: chain A and resseq 367 residue_selection_2: chain A and resseq 368 Model: "" Number of chains: 12 Chain: "A" Number of atoms: 7983 Number of conformers: 2 Conformer: "A" Number of residues, atoms: 496, 7926 Unusual residues: {'SNN': 1} Classifications: {'undetermined': 1, 'peptide': 495} Modifications used: {'COO': 1, 'NH3': 1} Link IDs: {'PTRANS': 37, 'TRANS': 456, None: 2} Not linked: pdbres="ASP A 366 " pdbres="SNN A 367 " Not linked: pdbres="SNN A 367 " pdbres="GLY A 368 " Chain breaks: 1 Unresolved chain link dihedrals: 1 Unresolved non-hydrogen bonds: 1 Unresolved non-hydrogen angles: 1 This fails with an error message that it cannot interpret the links. In my previous pdbs I have also defined the atoms which I wanted to link. Maybe I have to do this also somhow in Phenix. I have tried to extract the Link definition using phenix.link_edits but the resulting .edits file is empty. How can I resolve issue? Thanks Christian

2 1

Rebetta fragment library
by Wei Shi 10 Nov '11

10 Nov '11

Hi everyone, I am trying to generate a fragment library in Rebetta server for my sequence of interest to use with mr_rosetta. I can open the 9-mer and 3-mer fragment results: aat000_09_05.200_v1_3 and aat000_03_05.200_v1_3, but didn't see link to download these two files. Does anyone know how to download these two files? If I do copy and paste, what file type or format is compatible with mr_rosetta? Thank you so much! Best, Wei

2 2

define angle between groups of atom 'planes' for refinement restraint?
by Francis E Reyes 10 Nov '11

10 Nov '11

Hi all I was wondering if there was language in the monomer restraint libraries (.cif) or a custom phenix.refine geometry restraint a that defines an angle between two planes, where each plane is a group of coplanar atoms. Thanks! F --------------------------------------------- Francis E. Reyes M.Sc. 215 UCB University of Colorado at Boulder

5 4

nproc for autobuild
by Michael Thompson 09 Nov '11

09 Nov '11

Hello, I apologize if this question seems very basic, but I am new to phenix.autobuild. I would like to set up a series of phenix.autobuild runs on our cluster, and would like to maximize the speed. I am trying to make single rebuilt models using the command below: phenix.autobuild data=data.mtz model=coords.pdb map_file=map_coeffs.mtz multiple_models_number=1 n_cycle_rebuild_max=5 ncycle_refine=5 s_annealing=true refine_with_ncs=True place_waters=True nproc=?? run_command=/home/qsub.sh The AutoBuild documentation says "if you are using rebuild_in_place=False, then use nproc=4" and "if you are using rebuild_in_place=True, then use nproc=5. (Any more will not make any difference)." My interpretation of the documentation is that for the above AutoBuild run with multiple_models_number=1, I should set nproc=5 (or 4?). Is this correct? Also, if I were to change the command so that multiple_models_number=10, should I then use nproc=50 (i.e. 5 processors per model), or is it still the case that setting nproc>5 doesn't make a difference? Thanks in advance for your help! Mike -- Michael C. Thompson Graduate Student Biochemistry & Molecular Biology Division Department of Chemistry & Biochemistry University of California, Los Angeles miket(a)chem.ucla.edu

2 1

phenix.phaser output .param file
by G Y 08 Nov '11

08 Nov '11

Dear Experts, I run my script for phenix.phaser in bash shell and one of the output file is filename.param (please note there is no "s" at the end of param) There are only two lines inside this file: # [no title set] TNCS IGNORE ON I do not really know what this file doing for. And, if I run the phaser from GUI, I will not find any .param file. What is this file doing for? Is it some kind of redundant information which is not useful at all? What should I edit in input params.eff file to avoid this output file? I would appreciate to any help from you. Thanks. Best, G

3 2

Constrained group refinement
by Christian Roth 08 Nov '11

08 Nov '11

Dear all, I defined the following contrained groups which should a ligand in two alternative forms and an ion at the same position refine to an summed occupancy of 1. constrained_group { selection = chain Z and resname 2CL and altloc A selection = chain Z and resname 2CL and altloc B selection = chain r and resname Cl } constrained_group { selection = chain c and resname 2CL and altloc A selection = chain c and resname 2CL and altloc B selection = chain m and resname Cl } constrained_group { selection = chain N and resseq 69 and altloc A selection = chain N and resseq 69 and altloc B selection = chain o and resname 2Cl } } At the end I get an error because of the very small nonbonded interactions. it seems that my definition are not correct, but I am at the moment unsure why. Could someone point me to the mistake. Thanks in advace for your help Best Regards Christian

2 1

Constrained group refinement Addendum
by Christian Roth 08 Nov '11

08 Nov '11

Do the ions have to be altlocC for the group definition? Best Regards Christian

1 0

Cross Validation Reflection Set question
by Yuri 07 Nov '11

07 Nov '11

Thanks for the replies. >I just tried doing this, and it looks like the R-free flags will cover >the full resolution range, at least with the current code. By current code, you mean 1.7.2-869 and newer, right? >Is this really necessary in this case (refine against lower res data, >then continue with the whole data set I think I have a backbone shift in this structure (relative to native protein used for MR) so I wanted to do it this way befor trying to go all the way down to 2A. One last question does phnix do NCS averaging to improve maps? -- Yuri Pompeu

6 6

Re: [phenixbb] AutoBuild runtime
by Nathaniel Echols 07 Nov '11

07 Nov '11

On Mon, Nov 7, 2011 at 9:14 AM, Shya Biswas <shyabiswas(a)gmail.com> wrote: > Thanks for the suggestions. Unfortunately my MR solution is not good enough > to get a good map. Autobuild does give me a solution when i put in my > sequence but the map does not look good. I was wondering if the final model > that Autobuild gives has undergone density modification. Strictly speaking, it's the map that undergoes density modification - but yes, this will be done unless you explicitly disable it. -Nat

1 0

Cross Validation Reflection Set question
by Yuri 07 Nov '11

07 Nov '11

Dear All, Here is the situation, I have data to 2 A. I refined the model to 3 A first with S.A. Now I want to use higher res data for the rest of refinement. My Rfree created in the first round when I only used data to 3A. Was Rfree set created using all the data or just up to 3A shell? or do I have to extend into higher res shells? one caviot is I had twin law h,-k,-h-l enabled when they were created. Thanks, -- Yuri Pompeu

3 2