- Chimera-dev - RBVI Mailing Lists

Re: [chimera-dev] [Chimera-users] Import Chimera Modules
by Tom Goddard 18 Mar '13

18 Mar '13

Hi Jon, I'm switching this discussion to the chimera developer mailing list since it is so technical. I'm wondering why you want to run Chimera in your own Python. Often people want that so they can install their own custom Python modules and use them with Chimera. We recommend just using the Chimera Python for that and installing your other modules into the Chimera Python. That saves you a lot of grief trying to do what you did -- set up the environment to get Chimera to start in another Python. Tom -------- Original Message -------- Subject: Re: [Chimera-users] Import Chimera Modules From: Jonathan Williams To: Tom Goddard Date: 3/18/13 2:37 PM > Hi Tom, > > Good news: I got the chimera modules to import! > > I think that you were correct about having an older > libfreetype.6.dylib that was being used instead of Chimera's version, > probably from my Enthought Python Distribution . Overall I believe > that my EPD was making things harder for me. So I removed the EPD > from my system, set all of the environment variables per the Chimera > IDLE, added ~/Resources/lib and ~/Resources/share to PYTHONPATH, and > now I'm in business. > > Thank you for your help and suggestions! > > Jon > > > On Mar 18, 2013, at 12:49 PM, Tom Goddard <goddard(a)sonic.net > <mailto:goddard@sonic.net>> wrote: > >> Hi Jon, >> >> Chimera includes the version of libfreetype.6.dylib it needs. The >> error must mean you have another older libfreetype.6.dylib on your >> system that Chimera is getting instead of its own version. The first >> thing would be to try to locate where the other libfreetype.6.dylib >> is. I guess Chimera starts on your machine, so somehow it gets the >> right library. That suggests that you do not have the environment >> variables that control library loading setup in the same way as >> Chimera does. I'd suggest you start Chimera, open the IDLE shell >> window (Tools / General Controls) and print all the environment >> variables seen by the running Chimera using for example >> >> import os >> for k,v in os.environ.items(): >> print k,v >> >> and see if those differ from what you are trying outside Chimera. >> >> Tom >> >> >> >> -------- Original Message -------- >> Subject: Re: [Chimera-users] Import Chimera Modules >> From: Jonathan Williams >> To: Tom Goddard >> Date: 3/18/13 8:05 AM >>> Hi Tom, >>> >>> Thanks for the response. I set the DYLD_FALLBACK_LIBRARY_PATH >>> variable, and I think this took care of my original problem. >>> However I have encountered a new error: >>> >>> ImportError: >>> dlopen(/Applications/Chimera.app/Contents/Resources/lib/_chimera.so, >>> 2): Library not loaded: @executable_path/../lib/libfreetype.6.dylib >>> Referenced from: >>> /Applications/Chimera.app/Contents/Resources/lib/_chimera.dylib >>> Reason: Incompatible library version: _chimera.dylib requires >>> version 15.0.0 or later, but libfreetype.6.dylib provides version 13.0.0 >>> >>> This seems like I just need to update my FreeType libraries, but I >>> have not had a chance to try this yet. If you have any other >>> suggestions, those would be appreciated, otherwise I'll send an >>> update when I get a chance. >>> >>> Thanks again, >>> >>> Jon >>> >>> >>> On Mar 15, 2013, at 7:02 PM, Tom Goddard <goddard(a)sonic.net >>> <mailto:goddard@sonic.net>> wrote: >>> >>>> Hi Jon, >>>> >>>> Maybe the trick is Chimera uses environment variable >>>> DYLD_FALLBACK_LIBRARY_PATH instead of LD_LIBRARY_PATH or >>>> DYLD_LIBRARY_PATH. Here's the Chimera Mac start-up C++ code that >>>> sets the environment variables >>>> >>>> http://plato.cgl.ucsf.edu/trac/chimera/browser/trunk/dist/mac/chimera-aqua.… >>>> >>>> and a comment on this code change says "Use >>>> DYLD_FALLBACK_LIBRARY_PATH instead of DYLD_LIBRARY_PATH, so >>>> system frameworks will get system versions of shared libraries." >>>> >>>> Tom >>>> >>>> >>>> >>>> On Mar 15, 2013, at 2:49 PM, Jonathan Williams wrote: >>>> >>>>> Hello, >>>>> >>>>> I am currently trying to import Chimera modules into a non-Chimera >>>>> python installation. I have referenced previous messages from >>>>> this mailing list >>>>> (http://www.cgl.ucsf.edu/pipermail/chimera-users/2009-March/003620.html) >>>>> and have added the relevant paths to Chimera in the >>>>> LD_LIBRARY_PATH, PYTHONPATH, and CHIMERA environment variables. >>>>> >>>>> However when I try to import Chimera, Midas or "import >>>>> chimeraInit" followed by "chimeraInit.init([])" I receive the >>>>> following error: >>>>> >>>>> >>> import chimeraInit >>>>> >>> chimeraInit.init([]) >>>>> Traceback (most recent call last): >>>>> File "<stdin>", line 1, in <module> >>>>> File >>>>> "/Applications/Chimera.app/Contents/Resources/share/chimeraInit.py", >>>>> line 594, in init >>>>> import chimera >>>>> File >>>>> "/Applications/Chimera.app/Contents/Resources/share/chimera/__init__.py", >>>>> line 16, in <module> >>>>> from _chimera import BBox, Camera, Color, ColorGroup, >>>>> DirectionalLight, LODControl, Lens, LensViewer, Light, Material, >>>>> MaterialColor, Model, NoGuiViewer, OGLFont, OSLAbbreviation, >>>>> OpenModels, OpenState, PathFinder, PixelMap, Plane, Point, >>>>> PositionalLight, Selectable, SharedState, SpotLight, Sphere, >>>>> Texture, TextureColor, TrackChanges, Vector, Viewer, X3DScene, Xform >>>>> ImportError: >>>>> dlopen(/Applications/Chimera.app/Contents/Resources/lib/_chimera.so, >>>>> 2): Symbol not found: __cg_TIFFClientOpen >>>>> Referenced from: >>>>> /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO >>>>> Expected in: >>>>> /Applications/Chimera.app/Contents/Resources/lib/libTIFF.dylib >>>>> in /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO >>>>> >>> >>>>> >>>>> I am using MacOSX 10.8.2, Python 2.7.3, and Chimera 1.7 (build >>>>> 38197) for 64bit Mac. >>>>> >>>>> Any help or advice would be greatly appreciated, >>>>> >>>>> Jon Williams >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> Chimera-users mailing list >>>>> Chimera-users(a)cgl.ucsf.edu <mailto:Chimera-users@cgl.ucsf.edu> >>>>> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >>>> >>> >> > > > > _______________________________________________ > Chimera-users mailing list > Chimera-users(a)cgl.ucsf.edu > http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users

1 0

Following on new atom types (again)
by Elisabeth Ortega 18 Mar '13

18 Mar '13

Dear Chimera dev. Team, I'm trying to do the same as JD. The message that I receive is when I try to minimize is "Element Pt (atom #0.1.A@PT1) is not currently supported". As Mr Pettersen told me before, I defined the parameters of PT in heme-iron.frcmod file in ../bin/amber/dat/leap/parms and I also defined the "pt" atom in the MMTK Database. Then I added the charges without the covalent bond between the PT and its neighbours and finally I try to minimize with and without the bonds. Nothing happens. Probably I'm modifying the wrong files... Then, I have other question to you regarding the IDAM Type. When I minimize, the N3 atoms become N3+. How can I change it? Thank you so much for your time and your patience. -- Elisabeth Ortega Carrasco, Ph.D. Student Molecular Modelling of Transition Metal Systems Group Lab. C7/153, Chemistry Department Universitat Autonoma de Barcelona Tel.: +34935812857 email.: ortega.elisabeth(a)qf.uab.cat

3 2

Following on new atom types.
by Jean-Didier Maréchal 15 Mar '13

15 Mar '13

Hi, I am following on Elisabeth's question on the 23rd of january regarding setting up new atoms. I have added platinum (for cisplatinum) as a new atom type and proceed as you recommended (calculate the charge on isolated species and generate the covalent bond for the first coordination sphere of the metal). I have though two questions: 1. when running the minimization I have the following message: Atom whose IDATM type has no equivalent Sybyl type: #-2147483648.-2147483648:1@PT1-CPT (type: Pt) Although the minimization still proceed, I'd like to do things right. Could you tell me where I could change the sybyl atomtype. I looked with the idle the idatmType of this atom and it gives me Pt+. I don't understand what is wrong. 2. I'd like to rapidly change the charges of the cisplatinum using charges calculated with ab initio calculations. I was thinking to change these values from the idle changing the charge attribute previous to minimization. Does it sound correct? What would be the easiest way you would reckon? Best regards, JD **************************** Dr. Jean-Didier Maréchal Associate professor The Computational Biotechnological Chemistry Team Unitat de Química Física Departament de Química Universitat Autònoma de Barcelona Edifici C.n. 08193 Cerdanyola (Barcelona) Tel: +34.935814936 e-mail: JeanDidier.Marechal(a)uab.es www: http://g <http://asklipio.qf.uab.es>ent.uab.cat/jdidier<http://ent.uab.es/jdidier>

3 3

Chimera binary code issue
by jupiter 13 Mar '13

13 Mar '13

Hi, I have been reading Chimera document on Chimera web site and I am understanding that Chimera is better to run with hardware accelerator, but we don't have GPUs in our virtual machines, we have to use mesa and we are using xlib-glx driver for remote VNC connection. One major issue I am currently investigating is that the binary Chimera code can perform 3D rotation in mesa DRI driver, but very poor performance when using mesa xlib-glx driver on CentOS 6. Please correct me my following assumptions and options to rectify the problem: (1) Binary code could be hard linked with mesa DRI driver, the solution is to remove the hard link, or replace some libraries. (2) Chimera may not be able to run on mesa xlib-glx driver at all, no possible workarounds, no solution. (3) Chimera can be used with xlib-glx driver if we can build Chimera source code with mesa xlib-glx driver. I understand you don't encourage people to build Chimera from source code, nor do I wish. But if that is the only choice which can fix the problem, would you please point me instructions and procedures how to build the source code on Linux? I've looked at README.bootstrap from svn source code, but Thank you. Kind regards. Jupiter

3 7

neighbors
by Elisabeth Ortega 12 Mar '13

12 Mar '13

Dear Chimera Dev. Team, There is any way to select the neighbors of a selected atom without drawing the bond? If it is not possible... could you tell me how draw and erase bonds using a python function? Thank you for your time. -- Elisabeth Ortega Carrasco, Ph.D. Student Molecular Modelling of Transition Metal Systems Group Lab. C7/153, Chemistry Department Universitat Autonoma de Barcelona Tel.: +34935812857 email.: ortega.elisabeth(a)qf.uab.cat

2 1

2-D projection of volume in Chimera
by Jesus Montoya 04 Mar '13

04 Mar '13

Hi, Is there a way to project a 3-D volume into a single 2-D image plane in Chimera? For example, I want to load a volume twice into Chimera and view the two copies side by side in an identical orientation (manually set by rotating the volumes with the mouse). Then, I want to view one of them as a 2-D projection along the axis that comes out of the screen. While a "solid" representation of the volume is similar, it is not exactly what I need... Thanks!

3 2

question
by Paola Peshkepija 26 Feb '13

26 Feb '13

Good afternoon, I am a new chimera user. I am using Chimera to look at the backbone dihedral angle variations for 28 GFP structures. I was wondering if there is a way I can look at the changes in dihedral angles using the 28 structures altogether rather than looking at each structure individually. Please, let me know. Thank you in advance for your help! Paola

2 1

Electrostatic coloring using Chimera
by Daniel Lee 19 Feb '13

19 Feb '13

To Chimera Developers, Hello, this is Daniel Lee from University of Utah Department of Pharmaceutics and Pharmaceutical Chemistry. I am currently working as a Teaching Assistant in a DNA/Protein modeling class for graduate students. We are using UCSF chimera 1.17 to explore molecular structures. I have a hard time trying to get the system to show electrostatic surface with colors matching to the electronegativity. I followed the tutorials however it seems really difficult to reproduce what the tutorial did. Surface/binding analysis did not had the preselected electrostatic potential file. I looked through the system files however did not find anything. Would you be able to assist me with this issue? The PDB model I was exploring was 2KTZ and 2KU0. I used the fetch function to download this file. Thank you, Daniel

2 1

Chimera Problem
by Neil Bugeja 19 Feb '13

19 Feb '13

Dear Sir/Madam I am using Chimera as part of my thesis. I need to find a homology model for the human muscarinic M1 receptor. However, it always tells me service is down or an error when loading from uniprot. Am I doing something wrong? The Uniprot ID is P11229 and I get fetch ID failed Thanks -- Neil Bugeja

2 1

Add new Amber parameters
by Elisabeth Ortega 23 Jan '13

23 Jan '13

Dear Chimera dev. team, I have to use my own set of Amber99 parameters because I'm working with transition metals. I'm testing if I can minimize a benchmark system that contains one copper atom coordinated to 6 water molecules. To do it, I don't use any script, I use directly the tool "minimize" present in Tools-> Structure Editing. To do so I edited the frcmod.ff99SB file with the atom, bond, dihedral... all the fields that frcmod.ff99SB ask to me. That didn't work (Element "Cu" is not currently supported by MMTK). Then I followed the instructions here: http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/minimize/minimize.… and I put a file called "cu" with the required data in: lib/python*/site-packages/MMTK/Database/Atoms I tried again, and the same problem. In the same webpage, there are instructions of how can we edit the Amber parameters files present in: bin/amber11/dat/leap/parm I put my parameters in frcmod.ff99SB and I also tried to put it in frcmod.ff03, but it fails again. How can I solve this problem and perform the minimizations? I'm using both windows(x64) and linux(x86) versions of Chimera 1.6.2 (build 36855) but I planned to finish the testing with Chimera 1.7 that contains by default the Amber12 package. At the bottom of the message there is the mol2 file and the parameters that I'm testing. Thanks in advance for your time. Elisabeth O.-C. My mol2 file is: @<TRIPOS>MOLECULE cuh2o.mol2 19 12 1 0 0 SMALL AMBER ff99SB @<TRIPOS>ATOM 1 CU 0.0581 0.0224 0.0049 CU 1 UNK 2.0000 2 OH -2.2040 0.0438 0.0019 O.3 1 UNK -0.8230 3 OH -0.0067 -0.2388 2.1919 O.3 1 UNK -0.8370 4 OH 0.0041 0.2849 -2.1820 O.3 1 UNK -0.8450 5 OH 2.2006 0.0021 0.0073 O.3 1 UNK -0.8270 6 OH -0.0248 -2.4186 -0.2871 O.3 1 UNK -0.8330 7 OH 0.0202 2.4645 0.2971 O.3 1 UNK -0.8420 8 H15 -2.6953 -0.0699 0.6640 H 1 UNK 0.4010 9 H16 -0.6665 -0.1584 2.6929 H 1 UNK 0.4170 10 H18 -0.6618 0.2534 -2.6804 H 1 UNK 0.4330 11 H19 2.6899 -0.0779 -0.6612 H 1 UNK 0.4240 12 H20 0.6422 -2.9163 -0.2996 H 1 UNK 0.4180 13 H14 0.6417 2.9769 0.0874 H 1 UNK 0.4200 14 H17 -2.6936 -0.1186 -0.6513 H 1 UNK 0.4180 15 H22 0.6497 -0.1530 2.6964 H 1 UNK 0.4220 16 H21 0.6504 0.1511 -2.6890 H 1 UNK 0.4130 17 H13 2.6872 -0.2103 0.6483 H 1 UNK 0.4060 18 H24 -0.6602 -2.9263 -0.1105 H 1 UNK 0.4160 19 H23 -0.6488 2.9572 0.3459 H 1 UNK 0.4200 @<TRIPOS>BOND 1 18 6 1 2 14 2 1 3 10 4 1 4 9 3 1 5 8 2 1 6 19 7 1 7 6 12 1 8 17 5 1 9 16 4 1 10 15 3 1 11 13 7 1 12 5 11 1 @<TRIPOS>SUBSTRUCTURE 1 UNK 1 RESIDUE 4 A UNK 0 ROOT And my initial trial of parameters that I write in the mod file is (I put the same than Fe atom parameters written in the file frcmod.heme_ff94 only to test if that works): MASS CU 63.55 BOND CU-OH 200.00 2.0 ANGLE CU-OH-HO 35.00 120.00 OH-CU-OH 50.00 90.00 DIHEDAL X-OH-CU-X 1 0.00 0.00 2.0 NONBON CU 1.2 0.05 0.00 -- Elisabeth Ortega Carrasco, Ph.D. Student Molecular Modelling of Transition Metal Systems Group Lab. C7/153, Chemistry Department Universitat Autonoma de Barcelona Tel.: +34935812857 email.: ortega.elisabeth(a)qf.uab.cat

3 2