- phenixbb - phenix-online.org

divergence of R and Rfree at the last step of refinement
by Edward A. Berry 04 Jul '13

04 Jul '13

I'm refining a structure with phenix-dev-1414 version, and R-free decreases until the last step where R-work drops sharply and R-free increases. Am I doing something wrong? R-factors, x-ray target values and norm of gradient of x-ray target stage r-work r-free xray_target_w xray_target_t 0 : 0.3068 0.3347 4.111964e+00 4.184576e+00 1_bss: 0.2296 0.2735 3.919523e+00 4.036648e+00 1_weight: 0.2296 0.2735 3.919523e+00 4.036648e+00 1_xyzrec: 0.2377 0.2725 3.938818e+00 4.036514e+00 1_adp: 0.2311 0.2707 3.918953e+00 4.031556e+00 1_occ: 0.2311 0.2707 3.918944e+00 4.031541e+00 end: 0.2187 0.2795 3.907777e+00 4.050961e+00 ////////////////////////////////////////////////////////////////// Looking in the log file, it seems that this happens at the stage of correcting the solvent mask: ============================= updating all scales ============================= start: r_work=0.3152 r_free=0.3389 (reset all scales to undefined) bulk-solvent and scaling: r_work=0.2324 r_free=0.2703 remove outliers: r_work=0.2324 r_free=0.2703 *-> correct solvent mask: r_work=0.2187 r_free=0.2795 <--*** final: r_work=0.2187 r_free=0.2795 overall anisotropic scale matrix: V0: 2.5130,-0.3626,1.3424,-0.7174,-0.0625,-2.3276 V1: -0.0348,-0.0556,-0.0105,-0.0091,0.0085,0.0052

2 2

Very low resolution data set
by Xiang Li 03 Jul '13

03 Jul '13

Hi everyone, I am working on a very low resolution data set (14 A). It is a designed DNA crystal and unit cell size & symmetry fit well with the assumption. When I do the refinement I found the Rfree is extreme high, ~0.49 while Rwork is about 0.4. But the model seems fits in well with the density map. I have collected different datasets, and it gives about the same MR result for different sets. I do not have any experience to work with low resolution data. Here I have a few questions: 1 What is the expected Rfree and Rwork for such low resolution data? I looked up PDB bank, and there is a 11.5A crystal with Rfree=~0.4. But not sure whether there is a kind of value for Rfree to consider the refinement works or not. 2 There might be overfit problem during the refinement. Currently what I did is to optimize the weight with PHENIX gui, add restraints between basepairs and NCS restraint. I tried to manually adjust the weight, but seems the optimized weight gives the best result. Are there other things I should try? 3 How to confirm that the result is real? There will be larger than usual error for such low resolution data, and what the supposed range should be? Thanks a lot for your suggestions and help! Sincerely, Xiang

3 2

omit map for protein-DNA complex
by Wei Shi 02 Jul '13

02 Jul '13

Hi guys, I've been trying to generate the omit map for a protein-DNA complex structure using Phenix. I don't know why it takes forever to run and didn't seem like to finish at all.... The resolution of the dataset is 3.2A. Omit map type: simulated annealing. Omit region: composite. And, for the corresponding apo-protein structure, the simulated annealing omit map was generated within 2 or 3 hours on the same computer cluster. Let me know whether you have any idea about what might go wrong and whether there is anything I could do.... Thank you so much! Best, Wei

2 2

Bias in Rfree
by "Weiergräber, Oliver H." 02 Jul '13

02 Jul '13

Hello, while refining a structure with NCS, I was re-considering this paper from Michael Chapman's group: Fabiola F, Korostelev A, Chapman MS. Bias in cross-validated free R factors: mitigation of the effects of non-crystallographic symmetry. Acta Crystallogr D Biol Crystallogr. 2006 Mar;62(Pt 3):227-38. In essence, they suggest to exclude from the working set reflections that are (1) related to a test reflection by NCS (2) adjacent to a test reflection and its NCS mates in reciprocal space They have implemented this strategy in a CNS module (R-free2005), but the software is no longer actively supported. Does anybody know if something similar has ever been implemented in other software packages such as phenix? Or am I missing any significant downsides of the approach - apart from a reduction of the working set? Best regards, Oliver ================================================ PD Dr. Oliver H. Weiergräber Institute of Complex Systems ICS-6: Structural Biochemistry Tel.: +49 2461 61-2028 Fax: +49 2461 61-9540 ================================================ ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Juelich GmbH 52425 Juelich Sitz der Gesellschaft: Juelich Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender), Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt, Prof. Dr. Sebastian M. Schmidt ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------

3 2

Error while updating water
by Subhani Bandara 02 Jul '13

02 Jul '13

Dear all, I am refining a protein structure at 2.5 A. After running some refinements with water update option on, I manually added water to some positions and ran without water update option as it removes those water from the sites and show green density around. Then after some refinements like this, I tried to use water update option to add water to some newly appeared density. When I check the option to update water using dev 1418 version it stops with the following error. Also when not using water update option, refinement statistics got worse when I used 1418 compared to 1.8.2-1309 with refinements using same inputs and options. R work-0.1988 compared to 0.1920 with 1.8.2-1309 Rfree-0.2623 compared to 0.2490 with 1.8.2-1309 Traceback: File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/cctbx_project/libtbx/thread_utils.py", line 236, in run return_value = self._target(self._args, self._kwargs, self._c) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/cctbx_project/libtbx/runtime_utils.py", line 94, in __call__ result = self.target() File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/cctbx_project/libtbx/runtime_utils.py", line 73, in __call__ result = self.run() File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/runtime.py", line 69, in run refine_object = self._run(args, callback) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/runtime.py", line 85, in _run call_back_handler=callback) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/command_line.py", line 161, in run call_back_handler=call_back_handler) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/driver.py", line 1630, in run ncs_manager = self.ncs_manager).result File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/strategies.py", line 155, in __init__ neutron_refinement2 = self.neutron_refinement2) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/task_table.py", line 26, in __init__ self.__call__() File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/task_table.py", line 36, in __call__ self.__getattribute__(method_name)() File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/task_table.py", line 106, in run_sol mc = self.macro_cycle) File "/Applications/PHENIX-dev-1418/Contents/phenix-dev-1418/phenix/phenix/refinement/task_table.py", line 79, in _need_to_run else: raise RuntimeError("Not a valid mode: %s"%mode) Thanks in advance for any help Subhani

4 4

automatic weight optimization
by Michael Thompson 01 Jul '13

01 Jul '13

Hello Phenix Folks, I have a few questions about the automatic weight optimization in phenix.refine. I have read the manual, but I feel like maybe I'm missing something. I apologize in advance if these are naive questions, but here goes. I ran phenix.refine with a .def file, in which I set the following parameters: target_weights { optimize_xyz_weight = True optimize_adp_weight = True wxu_scale = 2.72 } My goal was to optimize both xyz and adp weights. In the log file, I see something like this: ===================== xyz refinement (lbfgs minimization) ===================== R-FACTORS RMSD CLASH RAMA ROTA CBET WEIGHT TARGETS work free delta bonds angl data restr 21.06 20.38 -0.67 0.033 2.0 6.1 0.0 6.6 0 none 0.154 0.1685 19.79 21.26 1.47 0.012 1.3 7.6 0.0 5.8 0 2.494 0.131 0.0499 Legend: - first line corresponds to starting state (before refinement) - R-factors reported in percent - delta is Rfree-Rwork in percent - CLASH is all-atom Molprobity clashscore - ROTA is percent of side-chain rotamer outliers - RAMA is percent of Ramachandran plot outliers - CBET is number of Cbeta deviations - WEIGHT is relative weight between X-ray (or neutron) target and restraints - TARGETS: the values of X-ray (or neutron) and restraints target functions ================================ ADP refinement =============================== ----------Individual ADP refinement---------- R-FACTORS <Bi-Bj> <B> WEIGHT TARGETS work free delta data restr 19.79 21.26 1.47 1.909 20.062 0.661 0.131 19.80 21.28 1.49 1.761 20.011 0.007 0.130 19.77 21.41 1.64 1.838 19.884 0.030 0.125 20.30 22.63 2.34 2.369 19.168 0.122 0.111 20.17 22.82 2.64 2.886 18.929 0.243 0.107 20.07 23.04 2.97 3.771 18.737 0.365 0.104 20.04 23.14 3.10 4.106 18.751 0.486 0.103 20.00 23.19 3.19 4.496 18.794 0.608 0.103 19.98 23.23 3.25 4.846 18.851 0.729 0.102 19.97 23.27 3.29 5.115 18.940 0.851 0.102 19.95 23.30 3.36 5.420 18.988 0.972 0.101 19.94 23.33 3.39 5.579 19.008 1.094 0.101 19.93 23.29 3.36 5.598 18.917 1.215 0.101 max suggested <Bi-Bj> for this run: 10.00 max allowed Rfree-Rwork gap: 5.0 range of equivalent Rfree: 0.0 Best ADP weight: 0.007 Accepted refinement result: 19.80 21.28 1.49 1.761 20.011 0.007 0.130 The output kind of makes me feel like the adp weight is being optimized, but the xyz weight might not be. There is a list of statistics for varying values of the adp weight, but there is no similar list in the output from the xyz refinement stages. So my questions are: Is the xyz weight being optimized? Did I miss something in the input that is required to turn on xyz weight optimization? If the weight is indeed being optimized, then I must be misinterpreting the log file. How can I check to make sure the xyz weight optimization is running? Thanks for the help!! Mike Thompson -- 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

Format problem of restraint file
by Xiang Li 30 Jun '13

30 Jun '13

Hi everyone, I generated the basepairing restraint file by command phenix.secondary_structure_restraints MR.1.pdb > ss.eff . Then I opened the ss.eff file with gedit and edit the basepairing information. However, after I finished editing and saved the file, the refine program can not read the file. The error is "This application does not support the file type for ss.eff (format:txt)". I have attached a copy of the eff file. Do I have a way to solve the problem? Thanks for your help! Sincerely, Xiang

3 2

unwanted disulfids and geometry restraints
by Ursula Schulze-Gahmen 30 Jun '13

30 Jun '13

I have 2 questions: 1) I am using phenix.refine on the command line with a .eff file as input that was modified from previous GUI phenix runs. My structure has several Zn atoms and readyset creates a geometry_restraints file for the Zn interactions. I am not sure if phenix.refine gets these restraints when I run it from the command line. Do I need to read this file in some specific way? 2) The other question is: phenix refines several of the Cysteine neighbors in the Zn environment as disulfids, but they are not. How do I prevent this from happening? Ursula -- Ursula Schulze-Gahmen, Ph.D. Assistant Researcher UC Berkeley, QB3 356 Stanley Hall #3220 Berkeley, CA 94720-3220

5 10

Re: [phenixbb] what about heavy atom as model for MR
by zhaoy 30 Jun '13

30 Jun '13

Hi, It's my pleasure to recieved so much wonderful advice and I am trying according to the suggestions. I have used phenix.explore_metric_symmetry to see if there is any relationship between tetragonal and orthorhombic lattices. But the result suggests there is no relationship. In addition, I have some questions about the molecular replacement when density map entered as model. How to define clash between the model and How the value of RMS error of the density model impact the final MR result? Thank you! -- Yan Zhao, M.Phil. National Laboratory of Protein Sciences Institute of Biophysics Chinese Academy of Sciences 15 Datun Rd. Beijing, 100101 China Randy Read 写: Sorry, I'm coming a bit late to this thread, and most things I would have suggested have already been covered by Tom and Paul. Certainly there are examples in the literature of people using MR to place a heavy atom substructure against the anomalous differences, but as Tom pointed out there's the complication of making sure that you get the Se sites from one molecule and not from a combination of symmetry-related molecules, which probably won't have the same relative arrangement in the other crystal form. The one remaining thing I'd like to explore is whether you're sure that you've carried out the molecular replacement with density properly. This has become easier (with Tom's phenix.cut_out_density program), but it's still relatively complicated and error-prone. First, you have to be sure that you've cut out the density essentially for one molecule, and not a mixture of symmetry-related molecules. If the location and rough shape of the molecule isn't clear in your map, you may have to try things like cutting out spheres of density from different places in the unit cell and seeing which gives the clearest MR solution. Second, you have to make sure that you place the density in a reasonably large unit cell, 2.5-4 times the extent of the density that you've cut out, before using it as an MR model. Third, you have to choose an appropriate value for the effective RMS error of the density model. If your phases go to 4A, then a number like 1.5A to 2A might be appropriate. In our tests, the map can be pretty noisy but you can still get a very clear signal from the MR search. Of course, if the phase information is very weak, it still may not work even with an optimal envelope. Best wishes, Randy Read On 23 Jun 2013, at 15:56, "Terwilliger, Thomas C" <terwilliger(a)lanl.gov> wrote: You could try both the isomorphous and the anomalous differences. Each one has some useful information. You can even combine them but I'm guessing it won't help too much. Paul Adams' suggestion to check out related space groups is a very good one. Phenix.autosol only tries the opposite hand, so you have to try all the different screw axis possibilities that are not related by inversion. Another thing you could try is to repeat your MR with density from your map ( against the 2.6 A native) and get a long list of solutions. For each solution take the phases after MR and use them to calculate an anomalous difference Fourier for your semet P2221 data. Then notice if there are peaks in this map, and whether these peaks have the same relationship to the MR-placed density as you saw in your original map. If yes, you may have a solution. All the best, Tom T From: phenixbb-bounces(a)phenix-online.org [phenixbb-bounces(a)phenix-online.org] on behalf of zhaoy [zhaoy(a)moon.ibp.ac.cn] Sent: Sunday, June 23, 2013 3:32 PM To: PHENIX user mailing list Subject: Re: [phenixbb] what about heavy atom as model for MR Thank you very much. I will try as you said. The P2221 dataset's qulity are not very well. The points from diffraction image are not sharp. Besides, native dataset is 2.6A, setmet anomalous datasets is only 4A. So I have not obtained the solution with p2221datasets. There is only one molecular per AUS at p4322. So, from the initial map, I can confirm the heavy atom sites from the same molecular. Maybe I have a general idea about the method you said. But I still puzzled which F difference should I use, isomorphous or anomalous differences? Thanks! -- Yan Zhao, M.Phil. National Laboratory of Protein Sciences Institute of Biophysics Chinese Academy of Sciences 15 Datun Rd. Beijing, 100101 China Terwilliger, Thomas C 写: I'm not sure if there is a tool that will create the file you want for this. You could however just write a little script that reads in a .sca file and calculates F from I, subtract F+ - F- to get Dano, square that, write out Dano** 2 and sigma(Dano**2) as another .sca file. Then you could use this as input to molecular replacement with your sites from the P4322 dataset. There is an important caveat to this approach however: your sites from the P4322 have to be all part of the same molecule, and this might or might not happen automatically. If some sites are from one molecule and others from another molecule, then in the other crystal form their relationship won't be correct. If there is only one molecule in the au then this might or might not be a big problem. If there are more than one it is very likely to be a big problem. This would apply to the density search that you carried out as well. If your density is good enough to see where the molecule is, you might be able to figure out which of the symmetry-related sites to include in your MR search. Nevertheless as this is easy it seems like a fine thing to do. However I am wondering if it will work because your 2.6 A semet anomalous dataset did not yield a solution by itself, which is a little surprising... Also of course if your P4322 solution is not actually right or even just not close enough it would not work. -Tom T From: phenixbb-bounces(a)phenix-online.org [phenixbb-bounces(a)phenix-online.org] on behalf of 赵岩 [zhaoy(a)moon.ibp.ac.cn] Sent: Sunday, June 23, 2013 7:52 AM To: phenixbb(a)phenix-online.org Subject: [phenixbb] what about heavy atom as model for MR Hi everyone, The heavy atoms(Se-Met) were located at P4322 space group(4A). But the initial density map is so bad that I can not build the initial model. Meanwhile, I have another two datasets, native(2.6A) and se-met anomalous dataset with space group P2221. But these two datasets have no solution. I have tried the initial map from sp P4322 as a model to do molecular replacement. But no MR solution. So, I want to know how can I use the heavy atom as model to do MR with isomorphous or anomalous differences? Thanks! -- Yan Zhao, M.Phil. National Laboratory of Protein Sciences Institute of Biophysics Chinese Academy of Sciences 15 Datun Rd. Beijing, 100101 China _______________________________________________ phenixbb mailing list phenixbb(a)phenix-online.org http://phenix-online.org/mailman/listinfo/phenixbb ------ Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: + 44 1223 336500 Wellcome Trust/MRC Building Fax: + 44 1223 336827 Hills Road E-mail: rjr27(a)cam.ac.uk Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk

1 0

Is DEN refinement working?
by Wallen, Jamie 29 Jun '13

29 Jun '13

2 1