- phenixbb - phenix-online.org

Are sigma cutoffs for R-free reflections cheating?
by Joe Krahn 11 Dec '09

11 Dec '09

Using a sigma cutoff in refinement is almost always a bad idea. It appears that some people still use them. The problem is that sigma cutoffs almost always improve R-factors, because it increases the denominator in the error/average equation. Some people incorrectly think that the reduced R-factor means that the cutoff was an improvement in the structure quality. The result of a sigma cutoff is that the R-factor can be made to look better than it really is. I think that this should be prevented by never excluding any R-free reflections by any sort of cutoff criteria. That keeps the R-free value unbiased. If culling R-free is required for proper error analysis, or just because the culled data are almost certainly bogus values, then the un-culled R-free could be a separate value. Now days, most people use a sigma cutoff of zero, so it is normally not a big problem. However, it appears that PHENIX still throws out many reflections where Fobs==0, which can be a significant fraction in the last shell with anisotropic data. Unfortunately, the exclusion of weak reflections depends on how amplitudes were derived. If using CCP4 Truncate, those weak reflections will be inflated a bit to a non-zero value, and a zero-sigma cutoff will have a significantly different affect. Therefore, I think that the default should be to use reflections with Fobs==0, with SigFobs > 0 as the criterion for non-absent reflections in reflection files without a missing number flag (i.e. CNS format). Joe Krahn

6 12

cif_as_mtz : column labels
by Bryan W. Lepore 11 Dec '09

11 Dec '09

a .cif has a bunch of labels (see below). the output .mtz has three: FOBS, SIGFOBS, R-free-flags ... without getting anyones' hands dirty, are the F's merged from the four correct columns, viz. F_plus, F_minus, (and sigmas)? -bryan (for pdb code, pls. off-list) labels in .cif : _refln.F_meas_au _refln.F_meas_sigma_au _refln.F_calc _refln.phase_calc _refln.pdbx_F_plus _refln.pdbx_F_plus_sigma _refln.pdbx_F_minus _refln.pdbx_F_minus_sigma _refln.pdbx_HL_A_iso _refln.pdbx_HL_B_iso _refln.pdbx_HL_C_iso _refln.pdbx_HL_D_iso

2 1

Clearing old NCS group definitions
by John Pascal 09 Dec '09

09 Dec '09

Hello, I am using NCS restraints in phenix.refine, and I am defining the restraints in a parameter file that I read into the GUI interface. After modifying the ncs parameters file, I see in the log file that phenix.refine has used the updated NCS groups, but also retains the old restraints settings. Therefore, the number of NCS groups doubles each time I make an edit. Is there a way to erase the previous NCS restraints? I see that the restraints are accumulating in the project_data/*.eff files. Are those read in by default on consecutive runs? I am probably missing something obvious, but I would appreciate any advice. Thanks. Best, John -- John Pascal, PhD ph 215.503.4596 Assistant Professor fx 215.923.2117 Thomas Jefferson University Biochemistry & Molecular Biology 233 South 10th Street, BLSB 804 Philadelphia, Pennsylvania 19107 ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

2 1

Is it possible to do Molecular Replacement for synthesized "small" molecule X-ray data?
by Yves Wang 09 Dec '09

09 Dec '09

Hi PHENIX users, We've got a synthesized molecule, called Cucurbit[7]uril, molecular weight 1163 Da. We were able to grow crystals of cucurbit[7]uril with one of its ligand. This crystal diffracted to 1.5A at our X-ray machine. The dilemma is that, at this resolution, we can't use small molecule crystallography softwares to solve the structure (correct me if I'm wrong, we need at least 1A resolution?), while macromolecular X-ray softwares are really not designed for synthesized molecules. We have collected diffraction data of this crystal with decent quality. So I'm wondering if there's any way we can put a known crystal structure of cucurbit[7]uril and solve the structure of the complex. We have tried AutoMR but the solution is not very encouraging. Any suggestion will be gratefully appreciated. Best, -- Yi(Yves) Wang Department of Biochemistry Structural Biology & Biophysics Program Duke University BS: University of Science and Technology of China School of Life Sciences, National Laboratory for Physical Sciences at Microscale Tel: +1-919-236-3307 (Cell) +1-919-684-0235 (Lab 1) +1-919-660-1634 (Lab 2) Office: A20 LSRC / 5301 FFSC E-Mail: yves.wang(a)duke.edu Mail: Box 90317, Chemistry Department

3 2

Re: [phenixbb] Problem using MOSFLM-integrated MAD data
by Chen, Yu Wai 08 Dec '09

08 Dec '09

Dear All, Thanks to all those who have responded to my previous query. The problem was due to I used MOSFLM raw data without scaling or sorting. I put the MOSFLM data through SORTMTZ and SCALA and output as "unmerged intensities in SCALEPACK format" and that error I mentioned earlier in the BB has gone. SOLVE completed successfully with some possible solutions. But now I have another problem. When RESOLVE tries to run DM, it reported another problem which seems to indicate it cannot read the scala-produced file correctly. The last bit of the log file before AutoSol stopped is as follows. I also print out the first few lines of the scala-produced SCALEPACK-format data file at the bottom for your diagnosis. Can you advise me what went wrong (and how to fix it)? Thanks a lot. Regards. Wai ##################### AutoSol_build AutoSol Run 22 Mon Dec 7 17:07:41 2009 SOLUTION FOR MODEL-BUILDING: Solution 2 using HYSS on /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl_ano_1.sca and taking inverse. Dataset #1 SG="P 21 21 21" SOLUTION FOR MODEL-BUILDING: Solution 1 using HYSS on /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl_ano_1.sca. Dataset #1 SG="P 21 21 21" DM SOLUTION USED FOR MODEL-BUILDING 1:Solution 2 using HYSS on /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl_ano_1.sca and taking inverse. Dataset #1 SG="P 21 21 21" WARNING: As the experimental data have been corrected for anisotropy and no input_refinement_file is available, no refinement will be carried out. Model-building will continue without refinement. Trying to guess refinement file for later use from ['/home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl'] Choosing guess of refinement file for later use from all files: ['/home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl'] FILE TYPE scalepack_no_merge_original_index GUESS FILE TYPE MERGE TYPE sca unmerged TARGET LABELS ['I', 'SIGI'] CONTENTS: ['/home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl', 'sca', 'unmerged', 'P 2 2 2', None, None, ['I', 'SIGI']] File /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.hkl ******************************************************************************** Failed to carry out AutoSol_build: float argument required, not NoneType ******************************************************************************** ########################### SeM8/phnx> more sem8_peak.hkl 4 P222 1 0 0 0 1 0 0 0 1 0 0 0 -1 0 0 0 -1 0 0 0 1 0 0 0 1 0 0 0 -1 0 0 0 -1 0 0 0 -1 0 0 0 1 0 0 0 -1 0 0 0 0 -1 -2 0 1 2 2 0 0 1 608.6 133.4 0 -1 -2 0 1 2 179 0 0 1 660.0 138.4 0 1 -2 0 1 2 176 0 0 2 519.0 130.9 0 -1 -3 0 1 3 4 0 0 1 1409.6 239.7 0 -1 -3 0 1 3 179 0 0 1 1525.0 251.0 -- Yu Wai Chen, PhD........................................Lecturer King's College London, Randall Division +44-207-848-8206 New Hunt's House, Guy's Campus, London SE1 1UL, U.K.

2 1

Problem using MOSFLM-integrated MAD data
by Yu Wai Chen 07 Dec '09

07 Dec '09

> > Message: 3 > Date: Fri, 4 Dec 2009 15:54:13 -0800 > From: Nathaniel Echols <NEchols(a)lbl.gov> > To: PHENIX user mailing list <phenixbb(a)phenix-online.org> > Subject: Re: [phenixbb] Problem using MOSFLM-integrated MAD data > Message-ID: <7B4AA329-8759-4603-BC26-9FD0A9120FC2(a)lbl.gov> > Content-Type: text/plain; charset=us-ascii > > On Dec 4, 2009, at 3:44 PM, Yu Wai Chen wrote: > >> > .... >> > Unable to analyze /home/wai/dbe15_sh/SeM8/phnx/sem8_remo.mtz with xtriage >> > (a total of 3 instances of this) >> > . . . >> > Testing anisotropy in /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.mtz ... >> > Applying overall anisotropic B correction to /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.mtz to yield AutoSol_run_8_/TEMP0/ANISO_sem8_peak.mtz.sca >> > Overall B set to minimum of the trace of B values >> > Unable to get range of aniso B from xtriage output >> > Multiple equally suitable arrays of observed xray data found. >> > >> > wavelength { >> > data = sem8_peak.mtz >> > labels = "I, SIGI, M_ISYM" >> > } >> > > What happens when you specify the labels as simply "I SIGI"? > > The same errors resulted. Any advise please? Wai -- Yu Wai Chen, PhD........................................Lecturer King's College London, Randall Division +44-207-848-8206 New Hunt's House, Guy's Campus, London SE1 1UL, U.K.

1 0

Re: [phenixbb] P32 twin or p3221?
by lauren 06 Dec '09

06 Dec '09

Hi Peter, Thanks for your reply. >This data can be processed to P3 or P321 with the R-fac almost the same >After that, I used Bales to do the MR, two solutions were found in P32 >(dimmer) and one in P32 21 (monomer) >Then, I refine the structure using phenix.refine in both P32 and P32 21 sets, but the R/Rfree stopped at 0.19/0.25 and never goes down. >However, if I treat the P32 data as a twin (with twin law h,-h-k,-l ), the R/Rfree can down to 0.12/0.18. >Do an R vs R analysis by loading up a model as well in xtriage to get a clue about the extend of the pseudo symmetry. I do the R vs R analysis with P32 in xtriage, but the R value seems very close to zero. And the D_ncs is high (0.95) in resolution range 9.9841-4.2623, which may indicate an incorrect point group? The results are as follow. --------------------------------------------- Analysing possible twin law : h,-h-k,-l --------------------------------------------- mean |H| : 0.018 (0.50: untwinned; 0.0: 50% twinned) mean H^2 : 0.001 (0.33: untwinned; 0.0: 50% twinned) Estimation of twin fraction via mean |H|: 0.482 Estimation of twin fraction via cum. dist. of H: 0.487 Britton analyses Extrapolation performed on 0.47 < alpha < 0.495 Estimated twin fraction: 0.455 Correlation: 0.9777 R vs R statistic: R_abs_twin = <|I1-I2|>/<|I1+I2|> Lebedev, Vagin, Murshudov. Acta Cryst. (2006). D62, 83-95 R_abs_twin observed data : 0.022 R_abs_twin calculated data : 0.090 R_sq_twin = <(I1-I2)^2>/<(I1+I2)^2> R_sq_twin observed data : 0.001 R_sq_twin calculated data : 0.006 Perfoming correlation analyses The supplied calculated data are normalized and artificially twinned Subsequently a correlation with the observed data is computed. Results: Correlation : 0.832594802609 Estimated twin fraction : 0.49 Maximum Likelihood twin fraction determination Zwart, Read, Grosse-Kunstleve & Adams, to be published. The estimated twin fraction is equal to 0.478 Log[likelihood]: 78991.418 twin fraction: 0.443 D_ncs in resolution ranges: 9.9841 -- 4.2623 :: 0.950 4.2623 -- 3.4280 :: 0.694 3.4280 -- 3.0083 :: 0.655 3.0083 -- 2.7394 :: 0.885 2.7394 -- 2.5466 :: 0.853 2.5466 -- 2.3986 :: 0.824 2.3986 -- 2.2800 :: 0.817 The correlation of the calculated F^2 should be similar to the estimated values. Observed correlation between twin related, untwinned calculated F^2 in resolutiuon ranges, as well as ewstimates D_ncs^2 values: Bin d_max d_min CC_obs D_ncs^2 1) 9.9841 -- 4.2623 :: 0.981 0.902 2) 4.2623 -- 3.4280 :: 0.970 0.482 3) 3.4280 -- 3.0083 :: 0.958 0.429 4) 3.0083 -- 2.7394 :: 0.954 0.783 5) 2.7394 -- 2.5466 :: 0.964 0.728 6) 2.5466 -- 2.3986 :: 0.953 0.679 7) 2.3986 -- 2.2800 :: 0.963 0.667 Best Regards, Lauren

1 0

Re: [phenixbb] Phenix installation issue
by smoniot＠chemie.fu-berlin.de 05 Dec '09

05 Dec '09

Well, I've dowloaded the installer again and changed my use of tar from -xvf to -xf option. Then reran the installation It went through and everything seems to work now. The only thing is that i'm normally running bash shell and that phenix doesn't work so. sourcing the phenix_env does not work under bash (setenv) Did I miss something or switching to tcsh is necessary ? Thanks for your help installing Seb > just a thought: > > if /net/anaconda is an nfs-shared drive, and you are trying to sudo > something on it from another workstation, it will fail because sudo > grants administrative privileges on the workstation, not on the server > where drive is housed. > > On Fri, 2009-12-04 at 04:59 +0100, smoniot at chemie.fu-berlin.de wrote: > Hi, > > I'm trying to install phenix on my computer (Linux distribution Kubuntu 64 > bits (virtual machine but it should not be an issue)) but the installation > failed after some minutes ... > the last lines on the installation log file are : > > /bin/tar: ./build/intel-linux-2.6-x86_64/base/bin/wx-config : un lien > symbolique ne peut pas être créé vers > «/net/anaconda/scratch1/phenix/phenix-1.5-2/build/intel-linux-2.6-x86_64/base/lib/wx/config/gtk2-ansi-release-2.8»: > Opération non permise > > It's in french but basiccaly it can't create a symbolic link to > /net/anaconda.... not allowed. > > I don't get it since I did try both the sudo ./install command and the su > - root. > > If anyone experienced this and get to fix it I would be glad to know :) > > Thanks in advance > > Sebastien > _______________________________________________ > phenixbb mailing list > phenixbb at phenix-online.org > http://phenix-online.org/mailman/listinfo/phenixbb

2 1

Problem using MOSFLM-integrated MAD data
by Yu Wai Chen 05 Dec '09

05 Dec '09

Dear All, I am trying to use a set of MAD data integrated with MOSFLM in phenix. But when I ran autosol, I've got the following error in the log file: .... Unable to analyze /home/wai/dbe15_sh/SeM8/phnx/sem8_remo.mtz with xtriage (a total of 3 instances of this) ... AutoSol_setup_for_scaling AutoSol Run 8 Fri Dec 4 23:18:59 2009 Shifting to Automatic mode.... RESOLUTION 2.5 SOLVENT_FRACTION 0.54 RESIDUES 364 MAD-N-HA 14 NCS-COPIES 2 Setting up for scaling dataset resolution: 2.5 solvent_fraction: 0.54 Scoring criteria will be: ['SKEW', 'CORR_RMS'] Scoring method will be: BAYES-CC AutoSol_scale_and_analyze_mad AutoSol Run 8 Fri Dec 4 23:18:59 2009 scale_and_analyze_mad for dataset 1 file type :ccp4_unmerged ccp4_unmerged merge type :unmerged Testing anisotropy in /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.mtz ... Applying overall anisotropic B correction to /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.mtz to yield AutoSol_run_8_/TEMP0/ANISO_sem8_peak.mtz.sca Overall B set to minimum of the trace of B values Unable to get range of aniso B from xtriage output Multiple equally suitable arrays of observed xray data found. ******************************************************************************** Failed to carry out AutoSol_scale_and_analyze_mad: Sorry, there is something wrong with the file /home/wai/dbe15_sh/SeM8/phnx/sem8_peak.mtz ... xtriage cannot analyze it? ******************************************************************************** THIS IS MY parameters input to phenix.autosol: autosol { seq_file = dbe15.seq sites = 14 atom_type = Se crystal_info { resolution = 2.5 space_group = P212121 } wavelength { data = sem8_peak.mtz labels = "I, SIGI, M_ISYM" lambda = 0.9794 f_prime = -7.83 f_double_prime = 5.97 } wavelength { data = sem8_edge.mtz labels = "I, SIGI, M_ISYM" lambda = 0.9796 f_prime = -9.72 f_double_prime = 3.08 } wavelength { data = sem8_remo.mtz labels = "I, SIGI, M_ISYM" lambda = 09719 f_prime = -4.11 f_double_prime = 3.85 } } Can some guru help me? I can send the mtz file if needed. -- Yu Wai Chen, PhD........................................Lecturer King's College London, Randall Division +44-207-848-8206 New Hunt's House, Guy's Campus, London SE1 1UL, U.K.

2 1

Phenix installation issue
by smoniot＠chemie.fu-berlin.de 05 Dec '09

05 Dec '09

Hi, I'm trying to install phenix on my computer (Linux distribution Kubuntu 64 bits (virtual machine but it should not be an issue)) but the installation failed after some minutes ... the last lines on the installation log file are : /bin/tar: ./build/intel-linux-2.6-x86_64/base/bin/wx-config : un lien symbolique ne peut pas être créé vers «/net/anaconda/scratch1/phenix/phenix-1.5-2/build/intel-linux-2.6-x86_64/base/lib/wx/config/gtk2-ansi-release-2.8»: Opération non permise It's in french but basiccaly it can't create a symbolic link to /net/anaconda.... not allowed. I don't get it since I did try both the sudo ./install command and the su - root. If anyone experienced this and get to fix it I would be glad to know :) Thanks in advance Sebastien

3 3