- phenixbb - phenix-online.org

sugar behaving sourly
by Patrick Loll 05 May '10

05 May '10

Hi, I'm refining a structure with an unusual monosaccharide. I generated a library for this sugar from PRODRG, and it looks OK, at least to my eye. Certainly the obvious parameters like bond angle restraints seem normal. However, when I refine with phenix, the sugar's C6 methyl group gets pushed out of density and assumes a very unnatural angle. Now, it turns out that there are two copies of this molecule in the AU, and the other copy looks fine after refinement. Moreover, I can load my library into Coot, and when I use it to refine the sugar's position or regularize its geometry, it works just fine; and when I run phenix.pdbtools --geometry_regularization and supply my library, the resulting structure has nice geometry. These three observations lead me to believe that there's nothing inherently wrong with my library. There are no steric clashes that would tend to push this methyl group out of whack (but just to be sure, I removed all nearby atoms and repeated the refinement, with the same result). I should mention that this refinement is at ca. 1.8 Å, and the density for the sugar looks quite nice. Any suggestions? Cheers, Pat Loll --------------------------------------------------------------------------------------- Patrick J. Loll, Ph. D. Professor of Biochemistry & Molecular Biology Director, Biochemistry Graduate Program Drexel University College of Medicine Room 10-102 New College Building 245 N. 15th St., Mailstop 497 Philadelphia, PA 19102-1192 USA (215) 762-7706 pat.loll(a)drexelmed.edu

2 1

Input NCS operators for AutoBuild
by Mark A Saper 05 May '10

05 May '10

This questions may be best answered by Tom. Does anyone know how to generate a custom NCS definition file that AutoBuild will read from the refinement format? The parameter is input_ncs_file = I used phenix.simple_ncs_from_pdb to generate 3 ncs format files. But I could not figure out how to make the program use my definitions. In any case simple_ncs_from_pdb outputs 3 formats in ./phenix/tmp: simple_ncs_from_pdb.ncs simple_ncs_from_pdb.ncs_spec simple_ncs_from_pdb.resolve I tried the last one, but it has no residue or chain specification. Should I use the .ncs_spec? The first one is for phenix.refine. Thanks, Mark _________________________________ Mark A. Saper, Ph.D. Associate Professor of Biological Chemistry University of Michigan Biophysics, 3040 Chemistry Building 930 N University Ave Ann Arbor MI 48109-1055 U.S.A. saper(a)umich.edu phone (734) 764-3353 fax (734) 764-3323 http://www.biochem.med.umich.edu/?q=saper http://www.strucbio.umich.edu/

3 2

Re: [phenixbb] NCS problems using phenix GUI on a Mac
by Mark A Saper 05 May '10

05 May '10

John, I am having the exact same problem. I need the NCS parameters in a format suitable for AutoBuild which I think is different than phenix.refine. Are you trying to use phenix.simple_ncs_from_pdb routine in GUI? Let me know if you have made any progress. Thanks, Mark > Hello, > > I am having difficulties defining NCS parameters using the GUI > interface (Mac 10.4.11; phenix 1.6-289). > > Phenix will automatically select NCS restraint groups and use them > during refinement. But if I edit the automated selections to my > liking then click on Update and Save, the refinement runs with the > original parameters unedited. I have tried reading in a ncs.param > file, but it does not get read during the refinement run. Also, an > intermittent problem with this GUI is that I am not allowed to "Add" > a new restraint group. When I click this button, it just clears out > the selections on the last group, and does not add a new group. Any > ideas? > > 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 _________________________________ Mark A. Saper, Ph.D. Associate Professor of Biological Chemistry University of Michigan Biophysics, 3040 Chemistry Building 930 N University Ave Ann Arbor MI 48109-1055 U.S.A. saper(a)umich.edu phone (734) 764-3353 fax (734) 764-3323 http://www.biochem.med.umich.edu/?q=saper http://www.strucbio.umich.edu/

1 0

Job Opening for Industrial Post Doc in Xray Crystallography Software Development
by Lance Westerhoff 05 May '10

05 May '10

Hello All- We have an opening at QuantumBio Inc. for an Industrial Post Doc in Xray Crystallography Software Development. Below is the short job description. The position is based at QuantumBio Inc. headquarters in State College, PA (home of Penn State), and begins immediately. We expect that the successful candidate will have a good understanding of the innards of PHENIX (hence the present audience) and be willing to work with our software engineers as needed to integrate PHENIX and QuantumBio's linear scaling QM offerings and of course validate the resulting method both on proprietary and public data. A good understanding of C++ and/or Python will go a long way. We will be validating the software on-site at at least one of our pharma collaborators along the way, so this is also a good opportunity to get to know practitioners in the field. Contact: Dr. Lance M. Westerhoff, lance(a)quantumbioinc.com, 814-235-6908 Job Description: Beginning immediately: QuantumBio Inc., headquartered adjacent to The Pennsylvania State University in scenic State College, PA, is in need of a post doctoral computational Biochemist/Chemist or Xray Crystallographer interested in continuing his or her career within a dynamic, small-company environment. Under the direction of Dr. Lance M. Westerhoff, General Manager of QuantumBio Inc., the successful candidate will work with both academic and industrial partners to help transition and validate the novel, quantum mechanically based Xray refinement technologies recently developed in the academic laboratory of Dr. Kenneth M. Merz Jr. to QuantumBio’s product line. The successful candidate will be integral to this project and will focus his or her efforts on the research and software development aspects necessary to validate and deploy this novel technology across the pharmaceutical industry. The position requires the candidate to have a Ph.D. in Chemistry, Biochemistry, or related field and have extensive experience with PHENIX and/or other similar xray refinement packages along with an understanding of modern software development principles. Due to the limited number of new work visas available, only American citizens or candidates with valid work permits will be considered at this time. A willingness to publish, and experience with C++ and high performance programming is a plus. Opportunities for advancement may be available through additional SBIR/STTR-style projects, and industry-supported collaborations. Further, in addition to the research and development duties of the project, the position will afford the successful candidate with access to and experience with QuantumBio’s industry partners including a number of pharmaceutical, biotech, and computational firms. Interested candidates should submit at their earliest convenience their CV, one or two representative publications, and a cover letter capturing the following items: * Outline of the types of structures you have worked with and your experiences in the field. * Summary your method research and software development experiences to date. * Discussion of your level of experience with QM, semi-empirical QM, and molecular mechanics. * Also, will you require sponsorship to legally work for a for-profit company in the US and when will you be available? About QuantumBio: QuantumBio Inc is accelerating drug discovery efforts by providing pharmaceutical, biotechnology companies and life science research organizations with a next generation of Computer-Assisted Drug Design (CADD) and Computer-Assisted Molecular Modeling (CAMM) solutions based on the increased accuracy and power of quantum mechanics. These solutions include a number of options tailored to each customer’s specific requirements including software licensing, services, and customization. See http://www.quantumbioinc.com/ for more! Thank you for your time. Please feel free to contact me if you have any questions or comments about the position. -Lance ____________________ Lance M. Westerhoff, Ph.D. General Manager QuantumBio Inc. WWW: http://www.quantumbioinc.com Email: lance(a)quantumbioinc.com Phone: 814-235-6908 Fax: 814-235-6909 This message and any attachments are solely for the intended recipient and should be considered confidential. If you are not the intended recipient, please immediately and permanently delete.

1 0

Re: [phenixbb] AUTOSOL for MAD data with error message-'Need a symfile for solve'
by Thomas C. Terwilliger 05 May '10

05 May '10

Hi Peter, I am a little puzzled...I would have thought that it was fixed in phenix 1.6.1. Can you possibly send me your data and exact command so that I can fix this? I'm sorry about that! -Tom T >> I am also having this problem, has this been corrected? I am using phenix-1.6.1 through the phenix GUI. >> >> Cheers >> Peter Czabotar >> >> >>>Hi again Frank and Graeme and Mohan, >>> >>>This is now fixed and should be available in tomorrow's build of PHENIX. Most setting should now be available in Autosol (this bug only affected Autosol, not Autobuild or any other PHENIX routines). >>> >>>Thanks for pointing out the problem (and Mohan if this does not solve it in your case, please let me know and I'll continue to fix it). >>> >>>All the best, >>>Tom T >>> >>>> Hi Frank and Graeme and Mohan, >>>> >>>> Yes, P 2 21 21 should be allowed in autosol. I see that it doesn't work and I'll fix that. >>>> >>>> All the best, >>>> Tom T >>>> >>>> >>>>>> Um, this one has been hacked to death on ccp4bb: there is in fact nothing wrong or "non-standard" with that setting. If the program won't >>>>>> handle it, email the authors and ask them to fix it. >>>>>> >>>>>> Reindexing is an almighty pain in the butt. >>>>>> >>>>>> That said: it sounds like you may be pointing at an out-of-date symmetry operator file; I forget which environment variable it is, check the manual, but point it to an up-to-date ccp4 installation, and >>>>>> I >>>>>> suspect it will be fine. >>>>>> >>>>>> phx >> >> >> >> ______________________________________________________________________ The information in this email is confidential and intended solely for the addressee. >> You must not disclose, forward, print or use it without the permission of the sender. >> ______________________________________________________________________ _______________________________________________ >> phenixbb mailing list >> phenixbb(a)phenix-online.org >> http://phenix-online.org/mailman/listinfo/phenixbb >>

2 1

Autobuild confrusion...which files to use?
by Carl Carruthers 04 May '10

04 May '10

using the GUI, i ran AutoMR and got several files, then hit the "Run AutoBuild" button. When the AutoBuild panel came up, it automatically loaded three files from the AutoMR run: the original .mtz used in the AutoMR with Data Type as "X-ray data;" MR.mtz with Data Type as "Map Coeffs."; and MR.pdb with Data Type as "Start Model." my question is: AutoMR lists the file MR.mtz as the Data type "Phased Data." the AutoMR run also created a file named MR.MAP_coeffs.mtz with Data Type as "Map Coefficients." Why did AutoBuild load MR.mtz as the "Map Coeff" instead of the MR.MAP_coeffs.mtz file as the "Map Coeffs?" which should i use? also, thanks for a great seminar in Houston of friday. you guys are awesome.... --Carl

3 2

anisotropy and mtz edit
by Leonid Sazanov 04 May '10

04 May '10

Hi, with highly anisotropic data we use Mark Sawaya's anistropy correction server, which works very well for electron density calculations. For refinement, we would like to use original data, but still cut to different limits over a,b,c (like 3.9 3.4 3.2 A, for example). Is there any facility for that in Phenix, and if not, can it be implemented? I know anisotropic scaling itself is done in Phaser, but would be useful also to have a possibility to cut resolution limits differently in any mtz file. -- Dr. Leonid A. Sazanov Research group leader Medical Research Council Mitochondrial Biology Unit Wellcome Trust / MRC Building Hills Road Cambridge CB2 0XY WEB: www.mrc-mbu.cam.ac.uk Tel: +44-1223-252910 Fax: +44-1223-252915

2 1

AUTOSOL for MAD data with error message-'Need a symfile for solve'
by Peter Czabotar 04 May '10

04 May '10

I am also having this problem, has this been corrected? I am using phenix-1.6.1 through the phenix GUI. Cheers Peter Czabotar >Hi again Frank and Graeme and Mohan, > >This is now fixed and should be available in tomorrow's build of PHENIX. >Most setting should now be available in Autosol (this bug only affected >Autosol, not Autobuild or any other PHENIX routines). > >Thanks for pointing out the problem (and Mohan if this does not solve it >in your case, please let me know and I'll continue to fix it). > >All the best, >Tom T > >> Hi Frank and Graeme and Mohan, >> >> Yes, P 2 21 21 should be allowed in autosol. I see that it doesn't work >> and I'll fix that. >> >> All the best, >> Tom T >> >> >>>> Um, this one has been hacked to death on ccp4bb: there is in fact >>>> nothing wrong or "non-standard" with that setting. If the program >>>> won't >>>> handle it, email the authors and ask them to fix it. >>>> >>>> Reindexing is an almighty pain in the butt. >>>> >>>> That said: it sounds like you may be pointing at an out-of-date >>>> symmetry operator file; I forget which environment variable it is, >>>> check the manual, but point it to an up-to-date ccp4 installation, and >>>> I >>>> suspect it will be fine. >>>> >>>> phx ______________________________________________________________________ The information in this email is confidential and intended solely for the addressee. You must not disclose, forward, print or use it without the permission of the sender. ______________________________________________________________________

1 0

B factor over 300
by Daqi Tu 03 May '10

03 May '10

Hi, I refined my structure (of which the data is very anisotropic) using phenix.refine and with individual_sites, individual_adp, and tls options turned on. From the refined coordinates, the highest B-factor in the ANISO record (ie B_tls + B_individual) is 305. This number looks rather high. Is it something I should worry about? and is there an easy way to compute the individual B-factor component? Thanks. Daqi Tu Eck lab Dana-Farber Cancer Institute

2 1

Re: [phenixbb] Rigid body refinement question
by Ralf W. Grosse-Kunstleve 02 May '10

02 May '10

> Is it possible to specify individual residues of a chain to undergo rigid > refinement while the rest of the chain does regular refinement? No, sorry. Ralf

5 10