- phenixbb - phenix-online.org

Fwd: autobuild failure
by abhishek dey 29 Jul '10

29 Jul '10

---------- Forwarded message ---------- From: abhishek dey <41.abhishek(a)gmail.com> Date: 29 July 2010 21:27 Subject: autobuild failure To: phenixbb(a)phenix-online.org Dear all, I am facing some problem while running the Autobuild option for PHENIX-1.6-289 version. I uploaded the mtz file, sequence of my protein and the pdb file and started running AUTOBUILD. but the program stopped with the following error message SG: P 2 21 21 Cell: [49.539999999999999, 99.140000000000001, 144.41999999999999, 90.0, 90.0, 90.0] Inverse hand of space group: P 2 21 21 Creating sg entry from /data/dey/2779/rv2779p22121_sort2.mtz Unit cell: (49.54, 99.14, 144.42, 90, 90, 90) Space group: P 2 21 21 (No. 18) Space group name is: P 2 21 21 symbol is: p22121 **************************************** AutoBuild Input failed Need a symfile for solve .the working spacegroup is P 2 21 21 it will be helpful if someone give useful suggestion on this. thanks in advance -- Abhishek Dey

2 1

Re: [phenixbb] phenixbb Digest, Vol 56, Issue 30
by Leonid Flaks 29 Jul '10

29 Jul '10

> Message: 4 > Date: Tue, 27 Jul 2010 08:39:17 -0700 > From: Nathaniel Echols<nechols(a)lbl.gov> > To: PHENIX user mailing list<phenixbb(a)phenix-online.org> > Subject: Re: [phenixbb] 32-bit binaries for version 1.6.3-473 > Message-ID: > <AANLkTinSzHiX4z9VPsrbMLFY7xFuwqwkghzCzj2WPjmd(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > On Tue, Jul 27, 2010 at 8:15 AM, Leonid Flaks<flaks(a)bnl.gov> wrote: > > >> I am running phenix 1.6.2-432 on a set of systems running 32-bit version of >> fedora-12. The latest version of phenix (1.6.3-473) has binaries mostly for >> 64-bit OS. The only 32-bit OS I see is fedora-3. I would be happy to switch >> to 64-bit OS one day, but HKL2000 that runs on the same computers has a big >> warning about instability of it on the 64-bit (it is dated back to 2005!) >> Coot is also crashing on the the 64-bit test system, so I have to stay with >> 32-bit for now. >> Does anybody know if fc-3 version would run on f-12? >> >> > I suspect it will be okay, but I haven't tested it. It partly depends on > which features will be used - the GUI is twitchier about system libraries on > Linux. Fortunately, we also distribute all of the source code for Phenix, > so if the binary installer doesn't work, I'd recommend switching to the > source installer. Are these the same machines (and OS) that were used for > RapiData? > > Yes, same 'cafe' systems. They all run f-12-32bit. > Would it be possible to have 32-bit version built for f-12 or f-13? > >> > Possibly - we are probably going to pull some machines from our queueing > system because they have only 8 cores, and some of these will become build > machines for additional platforms (64-bit Ubuntu and CentOS are candidates). > Also, we now build a CentOS 5 (similar to FC6) installer - it didn't work > properly for 1.6.3, but I think I've fixed it now, and we're going to push > out a 1.6.4 very soon because of some unrelated problems in 1.6.3. > > -Nat > > And on top of that, I don't see coot with python binary available for f-12 anymore. I will be testing fc-3 version then. Leon

3 2

32-bit binaries for version 1.6.3-473
by Leonid Flaks 27 Jul '10

27 Jul '10

Hi! I am running phenix 1.6.2-432 on a set of systems running 32-bit version of fedora-12. The latest version of phenix (1.6.3-473) has binaries mostly for 64-bit OS. The only 32-bit OS I see is fedora-3. I would be happy to switch to 64-bit OS one day, but HKL2000 that runs on the same computers has a big warning about instability of it on the 64-bit (it is dated back to 2005!) Coot is also crashing on the the 64-bit test system, so I have to stay with 32-bit for now. Does anybody know if fc-3 version would run on f-12? Would it be possible to have 32-bit version built for f-12 or f-13? Thanks, Leon

2 1

LigandFit parameters?
by Hermella Woldemdihin 27 Jul '10

27 Jul '10

Hi, I am using LigandFit on the GUI version of Phenix. In my electron density map there is a good density for my ligand. And I have my ligand in pdb format. As an input to LigandFit I gave my ligand file(pdb), an MTZ file from my data processing and my protein at some stage of refinement (@ a good R-values). I used the default settings (Resolution=0, Minimum CC of ligand to map=0.75 and Ligand map shape=Fo-Fc) and I got a bad result. My overall CC=0.33 and Score=68.68). What shall I do to get the best solution??? Thanks all! Hermi

2 1

Planarity restraint for covalent linkage
by Elwood A. Mullins 24 Jul '10

24 Jul '10

I'm trying to refine a thioester linkage (CoA to glutamate). Is it possible to define planarity restraints in a refinement.geometry_restraints.edits file along with bonds and angles? If not, what would be the best way to restrain the linkage?

3 3

questions about ncs copy in autobuild
by J J 23 Jul '10

23 Jul '10

Dear All, I am using phenix.autobuild .eff file to build a structure (version 1.6.1). For the input I have a pdb model file with 6 chains; a sequence file which contains residues of a single chain (the sequence file has a few more residues at N-terminal than the pdb model file); an mtz amplitude file; I also specified ncs_copies=6 (I know there are 6 copies in the au). What confuses me is the output pdb file only has one chain (huge Rwork&Rfree, >0.5). Should there be 6 chains? The other issue is that the output chain has fewer residues than my sequence file (both N&C-terminal). It also has fewer residues compared with the pdb model file. Is this because I do not have enough resolution to resolve those missing residues? Thanks in advance. Jason U Pitt

5 5

Defining heavy atom sites for AutoSol
by Kyle Dolan 23 Jul '10

23 Jul '10

Hi all, I used AutoSol to generate SAD phases for a protein after completing a partial MR model. My strategy worked, but I noticed there were more heavy atom sites in the solution than I was expecting. Having continued to build and refine the structure, I am fairly certain that I know which sites are the real sites and which ones are noise. I would like to use the real sites to reproduce the experimental phases in AutoSol or another Phenix module to perhaps improve the map quality (essentially running just SOLVE or Phaser without HYSS or model-building). What is the right approach? I am guessing something like this: phenix.autosol data.sca sequence.dat sites_file=sites.pdb build=False Cheers, Kyle Kyle T. Dolan Department of Biochemistry and Molecular Biology The University of Chicago ktd(a)uchicago.edu

3 2

Re: [phenixbb] phenix.automr ensemble/component question
by Thomas C. Terwilliger 23 Jul '10

23 Jul '10

Hi Scott, Your first command looks good to me. It is saying that the composition of the crystal is 2 copies of each chain: component_1.component_copies=2 component_2.component_copies=2 and you want to search for 2 copies of each of your search models: ensemble_1.copies_to_find=2 ensemble_2.copies_to_find=2 All the best, Tom T > Hi Phenixers, > > I'm a bit confused about the proper syntax for my phenix.automr job. > This is probably a simple concept, but I'm running on 3 hours of sleep > as my kid was up all night, and my synapses aren't working so good. > > I expect 2 copies of my protein per AU > > My protein is composed of two domains > > I have a single PDB search model for each domain > > is either of these commands correct? > > phenix.automr data=mydata.sca \ > ensemble_1.coords=domain1.pdb \ > ensemble_1.identity=95 \ > ensemble_1.copies_to_find=2 \ > component_1.mass=40579 \ > component_1.component_copies=2 \ > ensemble_2.coords=domain2.pdb \ > ensemble_2.identity=95 \ > ensemble_2.copies_to_find=2 \ > component_2.mass=13410 \ > component_2.component_copies=2 > > > or > > > phenix.automr data=mydata.sca \ > ensemble_1.coords=domain1.pdb \ > ensemble_1.identity=95 \ > ensemble_1.copies_to_find=1 \ > component_1.mass=40579 \ > component_1.component_copies=2 \ > ensemble_2.coords=domain2.pdb \ > ensemble_2.identity=95 \ > ensemble_2.copies_to_find=1 \ > component_2.mass=13410 \ > component_2.component_copies=2 > > > essentially I'm not really using an ensemble of search models, I only > have a single PDB for each domain. So should ensemble_1.copies_to_find > and ensemble_2.copies_to_find be set to 1 or 2? > > > Thanks, > Scott > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Scott Classen, Ph.D. > SIBYLS Beamline 12.3.1 > http://bl1231.als.lbl.gov > Advanced Light Source > Lawrence Berkeley National Laboratory > 1 Cyclotron Rd > MS6R2100 > Berkeley, CA 94720 > C) 510.206.4418 > O) 510.495.2697 > Beamline) 510.495.2134 > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > > > > _______________________________________________ > phenixbb mailing list > phenixbb(a)phenix-online.org > http://www.phenix-online.org/mailman/listinfo/phenixbb >

2 2

writing composite omit maps in .xplor
by Susan Tsutakawa 23 Jul '10

23 Jul '10

Hi all, Continuing the conversation on looking at Phenix maps in pymol, what is the best way to generate .xplor maps in Autobuild. I would like to generate a figure in pymol with an omit map. I wasn't able to find documentation explicitly stating how this can be done. Sincerely, Susan

6 7

Questions about sequence file vs pdb model file in Autobuild
by J J 22 Jul '10

22 Jul '10

Dear All, Facts: -input: a sequence file, a pdb model file, an mtz file; the sequence file is the same as the pdb model except that the sequence file has 6 more residues at N-terminal; the pdb model is almost the same as my protein structure. (In a word, I just want to add a few more residues to the current pdb model. Of course I can do it in coot by manually placing residues at the N-terminal, but why not autobuild?) -problem: the output chain sequence is not the same as the sequence file (not the same as the input model file either). It misses some residues at the N-terminal and it adds some residues at the C-terminal (but disconnected from the C-terminal). Is it normal in autobuild? How to correct it? Thanks. Jason U Pitt

2 2