- phenixbb - phenix-online.org

Re: [phenixbb] How to extend R-free set to higher resolution by new phenix-1.5.2
by Thomas C. Terwilliger 14 Dec '09

14 Dec '09

Hi PC, >> >> Hi, >> >> I know this topic was discussed before, but I am still confused and still >> unable to refine my map merged from hires data because phenix.refine >> complains "No array of R-free flags found" >> >> The command was used to generate a hires map. >> phenix.autobuild data=resolve_1.mtz hires_file=hires.sca >> \seq_file=sequence.dat maps_only=True >> >> So, now I have "exptl_fobs_phases_freeR_flags.mtz", "cycle_best_1.mtz", >> "overall_best_denmod_map_coeffs.mtz"...several files in my autobuild >> folder, What this command has done is to take the phases in resolve_1.mtz and combined them with amplitudes from hires.sca and added free R flags to generate exptl_fobs_phases_freeR_flags.mtz . Additionally the file overall_best_denmod_map_coeffs.mtz now contains the above amplitudes along with some kind of density-modified phases. I would caution that this is probably not what you really wanted to do. Instead: Almost always you should use for "data=" a file that contains experimental phases only (not density-modified phases). So it would be better to use the command: phenix.autobuild data=phaser_1.mtz hires_file=hires.sca \ eq_file=sequence.dat maps_only=True which would instead density modify using hl coeffs from modify phaser_1.mtz, extending phase information to the data in hires.sca . Now you will have exptl_fobs_phases_freeR_flags.mtz with (a) experimental amplitudes from hires.sca (b) experimental phases and hl coefficients (where available) from phaser_1.mtz (c) free R flags to the full resolution of hires.sca and overall_best_denmod_map_coeffs.mtz with coefficients for a density-modified map at the resolution of hires.sca. >> >> How do I merge low and high resolution data by >> "phenix.reflection_file_editor"? What files (input and output arrays) >> should I choose? I think reflection_file_editor will only copy arrays (or columns) of data, and not merge several columns of data into one. Note that merging high and low resolution data requires (1) a scale factor between the datasets and (2) possibly a b-factor or anisotropic b-factor relating the two datasets. It cannot be properly done simply by merging reflection files. The reflection_flle_editor will add free R flags if they are missing or extend them for you if they are present only to low resolution, however. This allows you to take a low-res dataset with free R flags and a high-res dataset with no free R and to transfer the existing low-res free R flags to the high-res dataset and to extend them to full resolution. If you want, you can use phenix.solve to scale low and high resolution data by pretending that they are part of a mad dataset: make a file "scale.csh" like this, after editing to match your case: #!/bin/csh -f cp $PHENIX/solve_resolve/ext_ref_files/c2.sym . phenix.solve<<EOD resolution 20 3.0 cell 76 28 42 90 103 90 symfile c2.sym readdenzo ! readformatted/readdenzo/readtrek/readccp4_unmerged premerged ! premerged/ unmerged read_intensities ! read_intensities/read_amplitudes lambda 1 mad_atom c fprprv_mad 1. fprimv_mad 1. rawmadfile low.sca rawmadfile high.sca skip_solve sad EOD Then run with chmod +x scale.csh ./scale.csh Now solve_1.sca should have your scaled, merged data. I hope that helps! -Tom T

3 2

How to extend R-free set to higher resolution by new phenix-1.5.2
by Pei-Chun Lin 12 Dec '09

12 Dec '09

Hi, I know this topic was discussed before, but I am still confused and still unable to refine my map merged from hires data because phenix.refine complains "No array of R-free flags found" The command was used to generate a hires map. phenix.autobuild data=resolve_1.mtz hires_file=hires.sca \seq_file=sequence.dat maps_only=True So, now I have "exptl_fobs_phases_freeR_flags.mtz", "cycle_best_1.mtz", "overall_best_denmod_map_coeffs.mtz"...several files in my autobuild folder, How do I merge low and high resolution data by "phenix.reflection_file_editor"? What files (input and output arrays) should I choose? Detail description will be very appreciate. Thanks for answer my question. PC _________________________________________________________________ Windows Live: Make it easier for your friends to see what you’re up to on Facebook. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/so…

1 0

Side chain at special position
by Nate Nicely 11 Dec '09

11 Dec '09

I have a structure wherein the data show alternate conformations for a particular Trp side chain. The problem is that each of these conformations also occupies a special position: rotamer A clashes with rotamer A of the symmetry object, and similar for rotamer B. I'm attaching a screenshot wherein I tried to capture what's going on. The prerefined structure has a yellow carbon backbone, and you can see how refinement pushes rotamer A out of its rightful place in the map. I readjusted it and added the alternate conformation as seen in the green carbon backbone, but refinement after that just pushes apart both rotamers (not shown). I'd appreciate advice how to handle refining this structure in phenix. Thanks, Nate N

4 3

Re: [phenixbb] structural validation and deposit format
by Pavel Afonine 11 Dec '09

11 Dec '09

On 12/11/09 12:55 PM, rui wrote: > SO in the output pdb file, there are another set of lines start with > ANISOU, what are those, should I keep those? These are ADPs - Atomic Displacement Parameters, or B-factors. They are as important as atomic coordinates (almost) stored in ATOM records -:) > I turned on the optimze_wxc and optimize_wxu flags in the GUI, it does > give better R/Rfree. Do I need to turn on every time or just use the > optimized weights, where I can find those optimized weight? In log file? Optimization of weights is usually time consuming and the default weights are just good most of the time. So I usually recommend running weights optimization as the last step to tune up your model. Of course there might be exceptions. Since optimal weight may be different from run to run, they are not stored in .def file for the next runs. Pavel.

1 0

structural validation and deposit format
by rui 11 Dec '09

11 Dec '09

Hi, All, Sorry for this stupid question. When I tried to use phenix to validate the structure, it generate geometry outliers, model validation, real-space correlations and a phenix_map_coeffs.mtz file which I assume could be open in coot. How should I use this generated mtz file? When I open in coot, it tellls me core dump. I'm wondering if phenix will automatically do the corrections or should I look one by one in the log file and modify? It also generate a test_reduced.pdb, should I use this reduced pdb ( H added) for next round of refinement? Thanks. Is the format from phenix.refine ready for deposit?

3 2

refinement
by r n 11 Dec '09

11 Dec '09

Hi all I do have very low resolution data (around 4 ang), what are the efficient way of doing refinement, either rigid body alone or rigidbody and tls or individual with group_adp. I did both, but not much significant changes in Rfree stays around 48/50%. Also I do have other questions 1. While doing rigid body refinement, phenix complaint about the special position and could not perform rigid body refinement, whereas individual site refinement is working fine. I do have to delete the atom in special position for rigid body refinement? Thanks

8 20

How to reload a project into the gui
by Dirk 11 Dec '09

11 Dec '09

Hi, during a phenix refinement the gui froze and I had to kill the process. After restarting the gui the project I was working with was disappeared - all other projects are still there. Is there a way to reload the project into the gui? All files including the project folder still exists but are somehow not recognized by the gui. I'm using dev-249 64bit. Greetings Dirk

2 1

Re: [phenixbb] How to reload a project into the gui
by Thomas C. Terwilliger 10 Dec '09

10 Dec '09

Hi Dirk, Try phenix.refine --gui Refine_xx/refine_xx.eff where xx is the run number of your refinement. That should preload all the parameters you had set at least. All the best, Tom T >> Hi, >> >> during a phenix refinement the gui froze and I had to kill the >> process. After restarting the gui the project I was working with was >> disappeared - all other projects are still there. Is there a way to >> reload the project into the gui? All files including the project folder >> still exists but are somehow not recognized by the gui. >> I'm using dev-249 64bit. >> >> Greetings >> >> Dirk >> >> >> _______________________________________________ >> phenixbb mailing list >> phenixbb(a)phenix-online.org >> http://phenix-online.org/mailman/listinfo/phenixbb >>

1 0

Are sigma cutoffs for R-free reflections cheating?
by Joe Krahn 10 Dec '09

10 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 10 Dec '09

10 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