- phenixbb - phenix-online.org

Error: "cannot allocate memory"
by sujata halder 19 Aug '14

19 Aug '14

Hi, I am trying to refine a protein with 60 chains (a T=1 virus; 5-3-2 symmetry) and each chain has 520 amino acids. I am using version 1.9 (Release tag: 1692). The job stopped midway and gave the error below followed by "Cannot allocate memory". I have run a similar job (i.e using a similar sized protein) before on an older version (version 1.7) and that worked. Looking at the error I can't really figure out if the problem is related to installation or a missing module or something different about the code. I can share my input and output files off-list. Here is the error: ====================== XYZ individual (reciprocal space) ====================== R-FACTORS RMSD CLASH RAMA ROTA CBET WEIGHT TARGETS work free delta bonds angl data restr 28.80 29.29 0.49 0.016 1.6 11.9 0.8 7.4 60 none 0.093 160.8996 Traceback (most recent call last): File "/usr/local/phenix-1.9-1692/build/intel-linux-2.6-x86_64/../../phenix/phenix/command_line/refine.py", line 11, in <module> command_line.run(command_name="phenix.refine", args=sys.argv[1:]) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/command_line.py", line 163, in run call_back_handler=call_back_handler) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/driver.py", line 1577, in run reference_model_manager = self.reference_model_manager).result File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/strategies.py", line 183, in __init__ ion_placement_manager = ion_placement_manager) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/macro_cycle.py", line 114, in __init__ self.call(self.xyzrec, "XYZ individual (reciprocal space)") File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/macro_cycle.py", line 142, in call func(prefix = prefix) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/macro_cycle.py", line 374, in xyzrec prefix = prefix) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/xyz_reciprocal_space.py", line 204, in __init__ call_back=(len(trial_weights)==1)) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/xyz_reciprocal_space.py", line 314, in try_weight result = self.show(weight=wt, print_stats=print_stats, call_back=call_back) File "/usr/local/phenix-1.9-1692/phenix/phenix/refinement/xyz_reciprocal_space.py", line 365, in show molprobity_scores = self.main.use_molprobity) File "/usr/local/phenix-1.9-1692/cctbx_project/mmtbx/model.py", line 1341, in geometry_statistics molprobity_scores = molprobity_scores) File "/usr/local/phenix-1.9-1692/cctbx_project/mmtbx/model_statistics.py", line 99, in __init__ self.clashscore = clashscore(pdb_hierarchy=pdb_hierarchy).get_clashscore() File "/usr/local/phenix-1.9-1692/cctbx_project/mmtbx/validation/clashscore.py", line 133, in __init__ verbose=verbose) File "/usr/local/phenix-1.9-1692/cctbx_project/mmtbx/validation/clashscore.py", line 257, in __init__ self.run_probe_clashscore(self.h_pdb_string) File "/usr/local/phenix-1.9-1692/cctbx_project/mmtbx/validation/clashscore.py", line 341, in run_probe_clashscore stdin_lines=pdb_string).raise_if_errors().stdout_lines File "/usr/local/phenix-1.9-1692/cctbx_project/libtbx/easy_run.py", line 158, in __init__ close_fds=not subprocess.mswindows) File "/usr/local/phenix-1.9-1692/build/intel-linux-2.6-x86_64/base/lib/python2.7/subprocess.py", line 709, in __init__ errread, errwrite) File "/usr/local/phenix-1.9-1692/build/intel-linux-2.6-x86_64/base/lib/python2.7/subprocess.py", line 1222, in _execute_child self.pid = os.fork() OSError: [Errno 12] Cannot allocate memory ------------------------------------------------------------------------------------------------------------- Thanks, Sujata

2 1

Phenix.refine with Ir
by Chen Zhao 15 Aug '14

15 Aug '14

Dear all, I am sorry for bothering you with this dumb question, but it occurs to me that phenix cannot refine a model with Ir, giving the error message of atoms with unknown nonbonded energy type symbols. Since this is my first time really working on the refinement, I have no clue of how to fix it. Would anybody give me a hint? Thanks a lot in advance! Sincerely, Chen

2 1

maps for PDB without SF's
by PC 15 Aug '14

15 Aug '14

2 3

van der Wall's radii
by Rituparna Sengupta 15 Aug '14

15 Aug '14

Hi! I'm new to the list. This might be a naive question to ask, but I'm looking for information about the set of van der Waal's radii popular with most protein crystallographers. Is there a certain list that phenix uses for building from the x-ray structures. If so, where can I find it. I shall be grateful for any thoughts or inputs on this. Thanks, Ritu Rituparna Sengupta Graduate Student Biophysics Graduate Program Record Lab Department of Biochemistry Biochemical Sciences Building University of Wisconsin-Madison

3 2

Computational Crystallography Newsletter - Volume 5, Number 2
by Nigel Moriarty 13 Aug '14

13 Aug '14

I am pleased to announce the publication of the latest issue of the Computational Crystallography Newsletter: http://www.phenix-online.org/newsletter/ A listing of the articles and short communications is given below. Please note that the newsletter accepts articles of a general nature of interest to all crystallographers. Please send any articles to me at NWMoriarty(a)lbl.gov noting that there is a Word Template on the website to streamline production. Articles Details of the Conformation-Dependent Library <http://phenix-online.org/newsletter/CCN_2014_07.pdf#page=12> Short Communications Connectivity analysis tools in CCTBX <http://phenix-online.org/newsletter/CCN_2014_07.pdf#page=4> phenix.fab_elbow_angle: Fragment Antigen-Binding elbow angle calculation tool <http://phenix-online.org/newsletter/CCN_2014_07.pdf#page=8> Fitting tips #8 – Acetyl Groups are like Peptides: Planar and trans <http://phenix-online.org/newsletter/CCN_2014_07.pdf#page=2> Cheers Nigel -- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : NWMoriarty(a)LBL.gov Fax : 510-486-5909 Web : CCI.LBL.gov

1 0

PSDI 2014, Registration and Abstract Info
by Anna Gardberg 13 Aug '14

13 Aug '14

*[image: Inline image 1]PROTEIN STRUCTURE DETERMINATION IN INDUSTRY 2014* The 22nd PSDI will take place at the Hotel Miragem in Cascais, Portugal, a coastal town located 30 kilometers west of Lisbon, Portugal. The meeting will start with registration and an opening reception on Sunday, 2nd November. This will be followed by two and a half days of scientific program starting Monday morning and finishing at noon on Wednesday, 5th November. A half day excursion to Lisbon is scheduled as a part of the meeting. The scientific program will include the following sessions: · Drug Discovery Stories · Challenging Targets · Biophysics Supporting Structural Biology · Binding Kinetics in Drug Discovery · Structural Biology of Biopharmaceuticals Registration for participants and exhibitors is open at the website: www.psdi2014.org *Please note the deadline for registration and payment: 31 Aug 2014* If you are interested in speaking in any of the scientific sessions, please contact the organizers at psdi2014(a)merckgroup.com with an abstract. Abstracts for “Binding Kinetics in Drug Discovery” are particularly encouraged. Additional information concerning the conference and further updates can be found at the PSDI2014 website <http://www.psdi2014.org/>. Vendor/Sponsor opportunities are available. If you have any questions, please contact the organizers at psdi2014(a)merckgroup.com. Please feel free to forward this announcement to anyone who may be interested in participating. We look forward to a great conference this fall. With kind regards, The PSDI 2014 organizing committee: · Tiago Bandeiras, IBET, Portugal · Anna Gardberg, EMD Serono, USA · Theresa Johnson, EMD Serono, USA · Beate Matlok, Merck KGaA, Germany · Djordje Musil, Merck KGaA, Germany psdi2014(a)merckgroup.com http://www.psdi2014.org

1 0

random half data sets
by Michael Thompson 12 Aug '14

12 Aug '14

Hello All, I'm interested in doing some comparisons of random half data sets, inspired by statistics like CC1/2, CC*, etc. Does phenix contain some tool to split unmerged data into 2 random sets? Thought I would ask before trying to write my own script. Thanks, Mike -- Michael C. Thompson Postdoctoral Scholar Institute of Genomics and Proteomics University of California, Los Angeles miket(a)chem.ucla.edu

2 1

Alternate sequence
by Mullins, Elwood A. 12 Aug '14

12 Aug '14

Hello all, I'm having trouble refining a protein-DNA complex in which one of the DNA strands has a mixture of two different nucleotides at one position. Both are non-standard and require CIF files. I tried assigning each nucleotide the same chain identifier and the same residue number but unique alternate location identifiers. This didn't work. Phenix refined the model but only linked one of the nucleotides with the rest of the chain, pushing the 3' OH and 5' phosphate out of the density for the other nucleotide. Coot won't even open the model output by Phenix. I assume there's some way to do this. Thanks in advance, Elwood Elwood A. Mullins, Ph.D. <> elwood.mullins(a)vanderbilt.edu<mailto:[email protected]> <> tel 615.936.5234 <> fax 615.343.6707

2 1

Calculate F (model)
by PC 12 Aug '14

12 Aug '14

2 2

Re: [phenixbb] pdb_redo
by Nathaniel Echols 12 Aug '14

12 Aug '14

> > From: "Lenin Domínguez" <acheron24(a)gmail.com> > Subject: Re: [phenixbb] pdb_redo > I was wondering about PDB_REDO too. Your command line is priceless, Nat. > > I have one further question: Using phenix.fetch_pdb --maps shouldn't I be > able tro reproduce a figure from a paper showing the electron density of a > ligand? > Well, yes, usually. But: http://www.ncbi.nlm.nih.gov/pubmed/23385452 The one caveat is that there are a few legitimate manipulations that could make the figure look slightly different from how it appears in e.g. Coot: 1. Use of a finer FFT grid (d_min/5 or smaller, versus d_min/4 or d_min/3 which are the defaults for Phenix and Coot respectively) 2. Maximum entropy treatment (removes Fourier artifacts, extends coefficients to high resolution without actually adding information) 3. Map sharpening (2) and (3) should really be explicitly mentioned, however. And I have yet to see a case where these would make the difference between a ligand being detectable or not by eye. (Maximum entropy can sometimes improve the performance of LigandFit but the density is still pretty obvious to a human.) The maps from PDB_REDO shouldn't be hugely different from the maps that come out of phenix.maps, so looking at those results would be another control. I have tried using phenix.fetch_pdb, PDB_REDO (conservative and aggresive > refinement) as well as the data straight from EDS. > > Any thoughts? > Send me the PDB ID (and ligand name) off-list and I'll take a look. -Nat

1 0