Refine with refmac and see if the positive density shows up there too. If it does, then either there is something peculiar about your structure or both programs suffer from the same bug that somehow only shows up with your particular dataset. If not, then it is indeed phenix bug (still rather specific to your dataset). Are positive density peaks centered around the corresponding atoms? Not sure what you mean by saying that peaks "extend to 4 sigma". To verify your suspicion about ADP weights, consider manually resetting B-factor of, say, one of the offending sulfurs and calculating the corresponding map. On Mon, 2010-12-20 at 09:06 -0800, Joseph Noel wrote:

Hi Pavel,

I've refined individual ADPs (anisotropic for the protein and isotropic for solvent). Still quite a bit of positive density on methionine sulfurs and most carbonyls. They start around 7.5 sigma and extend to 4 sigma. There are no other Fo-Fc peaks appearing for solvent yet to be modeled, etc. so maybe I should play with ADP weights. All positive density is appearing on atoms with occupancies of 1 so not much more I can do to lower the positive density.

Joe ___________________________________________________________ Joseph P. Noel, Ph.D. Investigator, Howard Hughes Medical Institute Professor, The Jack H. Skirball Center for Chemical Biology and Proteomics The Salk Institute for Biological Studies 10010 North Torrey Pines Road La Jolla, CA 92037 USA

Phone: (858) 453-4100 extension 1442 Cell: (858) 349-4700 Fax: (858) 597-0855 E-mail: [email protected]

Web Site (Salk): http://www.salk.edu/faculty/faculty_details.php?id=37 Web Site (HHMI): http://hhmi.org/research/investigators/noel.html ___________________________________________________________

On Dec 19, 2010, at 5:12 PM, Pavel Afonine wrote:

...

Hi Joe,

at 1.45A resolution it is most likely the best to refine individual anisotropic ADPs using phenix.refine (anisotropic for macro-molecule and isotropic for the solvent). In that case the "local sphere restraints" are not used ("local sphere restraints" are used in individual isotropic ADP refinement only). All the relevant details are here:

see "On atomic displacement parameters (ADP) and their parametrization in PHENIX" article:

http://www.phenix-online.org/newsletter/

I'm almost sure that refining isotropic ADPs instead of anisotropic causes these residual densities around atoms. It's known effect and I recall seeing it in at least two papers.

Another things to check:

- is it Met, or Se-Met? - radiation damage? Try refining occupancies of S. Although you said it's not only S, so may be not.

Good luck! Pavel.

...

I am refining a very well ordered structure at 1.45 A and find that after doing just about everything including optimization of weights, etc that there are still areas of positive density residing over S atoms of Met, backbone atoms, etc. I have added hydrogens and used them during refinement as well. The maps are of a high quality and Free R factors are quite good, ~ 17%. Is there anything that can be played with more such as values used in the ADP restraints window to try and achieve a difference map with far fewer areas of significant positive density (all greater then or equal to 4 sigma)?

I am not sure exactly what effect the values of distance power, average power, etc will have on refinement. The other option I was thinking of was the scattering table options. I have other structures of this protein that extend to about 1.2.

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

-- "I'd jump in myself, if I weren't so good at whistling." Julian, King of Lemurs

I think Joe has figured it out already. Trying a different program when the one you are using gives you trouble is a sure way to see if the problem lies with data (I assume that buster, cns, phenix, refmac and shelx all produce maps that resemble the same reality, and if positive density comes from data it will hopefully show up in all cases). As for metaphors, it's more like plugging the same video card in two computers to see if the problem is the card itself or the motherboard. IIUC, the kick maps are intended to reduce bias and maybe get one a better density in not-so-clear region. As far as well refined model regions go, calculating phases from a set of models with randomly shifted atoms seems to certain extent resemble increasing the B-factors. So the positive density that shows up first on the heavier and/or less disordered atoms seems like an expected outcome. Cheers, Ed. On Mon, 2010-12-20 at 21:43 -0800, Pavel Afonine wrote:

...

Refine with refmac and see if the positive density shows up there too. If it does, then either there is something peculiar about your structure or both programs suffer from the same bug that somehow only shows up with your particular dataset. If not, then it is indeed phenix bug (still rather specific to your dataset).

Although this may definitely give more clues, I see this way of debugging this problem more like treating the appendicitis with Tylenol -:) This may help in some way but will unlikely identify the problem and fix its root. Solving a puzzle like this typically involves reducing the amount of "variables" by simplifying the task that still results in problem rather than making it more complex to solve by involving more "variables". Oh, I guess you know what I mean by "variables" so I don't go into it.

All the best! Pavel.

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

-- Edwin Pozharski, PhD, Assistant Professor University of Maryland, Baltimore ---------------------------------------------- When the Way is forgotten duty and justice appear; Then knowledge and wisdom are born along with hypocrisy. When harmonious relationships dissolve then respect and devotion arise; When a nation falls to chaos then loyalty and patriotism are born. ------------------------------ / Lao Tse /

Hi Joe, hmm... this is strange. I'm not in a convenient place to seriously look into this right now... but if you send me the data and model and tell which atoms are in problem, then I will have a look Thursday evening or Friday morning first thing (when I come back from vacation). I would like to know why kicked maps result in this effect and I want to fix this problem. Pavel. On 12/20/10 10:45 AM, Joseph Noel wrote:

Pavel et al.,

So it turns out I was generating "kicked" sigmaa weighted maps and not the traditional sigmaa weighted maps. When I return to the default mtz maps of phenix, then the positive density appearing on heavy atoms disappears. Has anyone seen similar results with kicked maps? One other thing, I have also unchecked the reject outliers option but in the results window of the GUI it still seems to be rejecting a few reflections. Is there a way to double check things to ensure no reflections are being rejected during refinements and map calculations?

Joe ___________________________________________________________ Joseph P. Noel, Ph.D. Investigator, Howard Hughes Medical Institute Professor, The Jack H. Skirball Center for Chemical Biology and Proteomics The Salk Institute for Biological Studies 10010 North Torrey Pines Road La Jolla, CA 92037 USA

Phone: (858) 453-4100 extension 1442 Cell: (858) 349-4700 Fax: (858) 597-0855 E-mail: [email protected]

Web Site (Salk): http://www.salk.edu/faculty/faculty_details.php?id=37 Web Site (HHMI): http://hhmi.org/research/investigators/noel.html ___________________________________________________________

On Dec 19, 2010, at 5:12 PM, Pavel Afonine wrote:

...

Hi Joe,

at 1.45A resolution it is most likely the best to refine individual anisotropic ADPs using phenix.refine (anisotropic for macro-molecule and isotropic for the solvent). In that case the "local sphere restraints" are not used ("local sphere restraints" are used in individual isotropic ADP refinement only). All the relevant details are here:

see "On atomic displacement parameters (ADP) and their parametrization in PHENIX" article:

http://www.phenix-online.org/newsletter/

I'm almost sure that refining isotropic ADPs instead of anisotropic causes these residual densities around atoms. It's known effect and I recall seeing it in at least two papers.

Another things to check:

- is it Met, or Se-Met? - radiation damage? Try refining occupancies of S. Although you said it's not only S, so may be not.

Good luck! Pavel.

...

I am refining a very well ordered structure at 1.45 A and find that after doing just about everything including optimization of weights, etc that there are still areas of positive density residing over S atoms of Met, backbone atoms, etc. I have added hydrogens and used them during refinement as well. The maps are of a high quality and Free R factors are quite good, ~ 17%. Is there anything that can be played with more such as values used in the ADP restraints window to try and achieve a difference map with far fewer areas of significant positive density (all greater then or equal to 4 sigma)?

I am not sure exactly what effect the values of distance power, average power, etc will have on refinement. The other option I was thinking of was the scattering table options. I have other structures of this protein that extend to about 1.2.

Hi Nat, I tried what you suggested (dragging my ss restraints file into the input area), and it does not seem to be working (i.e. my edited restraints do not show up; also the filename never appears in the box). I tried changing the fsuix to .eff or .def or .txt, none worked. If instead I drag my restraints file onto the phenix.refine icon, I get a warning about 'this application does not support the filetype ...'. I assume that as usual I am doing something wrong. Any ideas? Thanks, Luke Nathaniel Echols wrote:

On Mon, Dec 20, 2010 at 7:43 AM, Luke Rice wrote:

...

I am trying to evaluate whether or not secondary structure restraints are going to help me with the refinement I am working on. When I ran in 'auto' mode, I noticed that by default phenix specifies several a-helices as 3_10. So, I would like to use an edited list of secondary structure restraints.

Interesting, usually it's the reverse problem (3_10 helices annotated as alpha) - but the annotation program can be a little twitchy. I had intended for this to be graphically editable, but now that I've checked I see there's no way to actually change the helix type in the GUI. Will fix that soon.

...

Can I specify this file in the GUI? (Like Joe Noel, I, too, have become soft). I looked in what I thought to be the obvious places and did not turn anything up.

You can just drag it into the main input file list. Any parameter file may be used this way, with the caveat that the GUI will not actually display the parameters it contains (but they will be incorporated into the final config file). For this reason I recommend against using complete .eff or .def files from command line runs, but the output of phenix.secondary_structure_restraints (or the edited version thereof) is pretty safe.

Alternately, if you drag the parameter file onto the phenix.refine icon in the main GUI and release the mouse button, it should launch the phenix.refine GUI with all parameters in the file loaded in displayed in the GUI. (Since this isn't very obvious, I'll add a menu item somewhere to do the same thing less opaquely.)

-Nat _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

-- Luke M. Rice Assistant Professor and Thomas O. Hicks Scholar in Medical Research Department of Biochemistry, ND10.300 UT Southwestern Medical Center 5323 Harry Hines Blvd. Dallas, TX 75390-8816 phone: (214) 645-5931 email: [email protected]

To the PHENIXbb, [Phenix version 1.6.4-486] I'm using a .cif file for a custom iron-atom-containing prosthetic group that used to work with an older version of Phenix. Now I'm getting a "Conflicting scattering type symbols" error. Phenix.refine also halts with the following information: Initial symbol: "F" (from pdb element column) New symbol: "FE" (with residue name SF4) I've changed the element column in my .pdb file from "F" to "FE", but that did not help and while there was a previous post about this error on the PHENIXbb concerning TaBr, I have not found a solution. On a related note, are there now more options for explicitly defining the oxidation state of metal atoms in custom ligand/prosthetic groups (i.e. so the correct scattering factors are used)? Thanks for your help, -Andy Torelli

Hi Ralf, I believe correcting the spacing as you suggested did the trick. May I suggest changing the error message output slightly to read something like: [snip] Actual symbol: " F" (from pdb element column) Expected/New symbol: " FE" (with residue name SF4) [/snip] By including actual white spaces within quotations the error message will clue the user in to the fact that the spacing is off in their .pdb file (as it was for me). Maybe you could also include an explicit reminder that the spacing is critical and should be checked. Also, the Actual/Expected terminology seems more logical to me than Initial/New. Just a friendly suggestion. Thanks for your help and have a nice holiday, -Andy Torelli -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Ralf W. Grosse-Kunstleve Sent: Monday, December 20, 2010 6:38 PM To: PHENIX user mailing list Subject: Re: [phenixbb] How to resolve "Conflicting scattering type symbols" error Hi Andrew,

Initial symbol: "F" (from pdb element column)

New symbol: "FE" (with residue name SF4)

I've changed the element column in my .pdb file from "F" to "FE", but that did not help

Did you shift the "F" one column to the left in the PDB file when adding the "E"? The correct positioning is critical.

On a related note, are there now more options for explicitly defining the oxidation state of metal atoms in custom ligand/prosthetic groups (i.e. so the correct scattering factors are used)?

Yes, that should work now. Just add the charge after the element name in the PDB file. You can check in the phenix.refine output. Look for the output like this: Number of scattering types: 4 Type Number sf(0) Gaussians O2- 1 9.97 2 O 20 7.97 2 N 12 6.97 2 C 33 5.97 2 sf(0) = scattering factor at diffraction angle 0. Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb