- Chimera-users - RBVI Mailing Lists

Chimera release candidate
by Elaine Meng 24 May '07

24 May '07

Hello everybody, A new version of Chimera (1.2415) is available for all platforms: http://www.cgl.ucsf.edu/chimera/download.html It is a candidate for a production release we plan to make in about a week. Many new features have been added since our last production release (1.2304, October 2006), as detailed in the release notes: http://www.cgl.ucsf.edu/chimera/1.2415/docs/relnotes/snapshot.html Here are just a few of the changes since last October: - embedded POV-Ray for production of glossy, shadowed images including labels - ability to supersample or to use POV-Ray when recording movies - new input formats: MDL MOL/SDF, Gromacs trajectories - new Fetch by ID types: EDS (fo-fc) electron density difference map, PQS predicted biological unit - Presets menu with display settings for interactive Chimera use or generation of publication images - Side View includes "top view" for interactively adjusting stereo parameters and perspective - expanded morphing capabilities - new Fit Map in Map tool - new windowsize command - updates to handle new naming conventions in the remediated PDB Enjoy! (and do let us know right away via "Report a Bug" or chimera- bugs(a)cgl.ucsf.edu if you find any problems) Thanks, Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

1 0

Re: [Chimera-users] fit map in map
by Thomas Goddard 24 May '07

24 May '07

Hi Gabor, There are no Chimera commands to fit one map into another, but it can be done with a Python script. I've attached a Python script that I made by piecing together code from gui.py and __init__.py from directory chimera/share/FitMap. Tom Gabor Papai wrote: > Hi Tom, > > Can the fit map in map tool be used without the GUI and retreive back > only the transformation matrix? > Bye, > Gabor > # ----------------------------------------------------------------------------- # Example script for fitting one map in another without the graphical user # interface. # # chimera --nogui fitnogui.py # # It can also be run using the graphical Chimera interface using File/Open. # def fit_map_in_map(map1_path, map2_path, initial_map1_transform = None, map1_threshold = None, ijk_step_size_min = 0.01, # Grid index units ijk_step_size_max = 0.5, # Grid index units max_steps = 100, optimize_translation = True, optimize_rotation = True): # Files have to have file suffix indicating volume format. from VolumeData import open_file data1 = open_file(map1_path) data2 = open_file(map2_path) from VolumeViewer import Data_Region, full_region, Rendering_Options ro = Rendering_Options() map1 = Data_Region(data1, full_region(data1.size), '', ro) map2 = Data_Region(data2, full_region(data2.size), '', ro) if initial_map1_transform: from PDBmatrices.matrices import chimera_xform xf = chimera_xform(initial_map1_transform) map1.surface_model().openState.globalXform(xf) use_threshold = (map1_threshold != None) from FitMap import map_points_and_weights, motion_to_maximum points, point_weights = map_points_and_weights(map1, use_threshold) if len(points) == 0: if use_threshold: print 'No grid points above map threshold.' else: print 'Map has no non-zero values.' return move_tf, stats = motion_to_maximum(points, point_weights, map2, max_steps, ijk_step_size_min, ijk_step_size_max, optimize_translation, optimize_rotation) if initial_map1_transform: from PDBmatrices.matrices import multiply_matrices move_tf = multiply_matrices(move_tf, initial_map1_transform) header = ('\nFit map %s in map %s using %d points\n' % (map1.full_name(), map2.full_name(), stats['points']) + ' correlation = %.4g, overlap = %.4g\n' % (stats['correlation'], stats['overlap']) + ' steps = %d, shift = %.3g, angle = %.3g degrees\n' % (stats['steps'], stats['shift'], stats['angle'])) print header from FitMap.gui import transformation_description tfs = transformation_description(move_tf) print tfs # ----------------------------------------------------------------------------- # Example call fitting map into itself. # p1 = p2 = '/usr/local/src/chimera-demos/volume/examples/groel.mrc' fit_map_in_map(p1, p2)

1 0

Electrostatical surface of a virus
by Eduardo Sanz-Garcia 23 May '07

23 May '07

I need to visualize the electrostatic surface of a virus. I don't know how to do it. I was thinking in the following sequence of instructions: 1. Use the DelPhiController to generate the potential map (*.phi) for the asymmetric unit. 2. Use Multiscale Models tool to generate the 60-related units. Is this a logic sequence ? Any hint would be appreciated. Thank you.

2 2

Mac stereo
by Greg Couch 23 May '07

23 May '07

To deal with the issue that the Mac OS X's X11 doesn't work with stereo very well, we have decided to release an Aqua version of chimera concurrent with production releases of the X11 version. The Aqua version will be labeled as a snapshot because we still recommend using the X11 version (the user interface works much better than the Aqua version). But if you need stereo, multisampling (full window antialiasing), or background transparency/opacity, then you need to use the Aqua version. Greg Couch UCSF Computer Graphics Lab

1 0

Re: [Chimera-users] Virus Images for the MSD Database
by Tom Goddard 23 May '07

23 May '07

Hi Jawahar, Unfortunately Chimera is not able to save images without displaying the graphical user interface. The image saving code grabs the image directly from the main window OpenGL graphics buffer, so it also will not work if the main window is covered or iconified. Greg Couch once made a special version of Chimera for Linux using the OSMesa graphics library that could do all the graphics off-screen including saving images. But I don't see that version in the Chimera download archive, and I believe it was made more than a year ago. We recently added raytraced image capture using POVray. Since that does not use the OpenGL graphics it should be able to work in nogui mode but a Chimera bug report indicates that does not work http://www.cgl.ucsf.edu/cgi-bin/gnatsweb.pl?cmd=view&pr=3678&database=chime… and is expected to be difficult to fix. Unfortunately the only option at this time is to have the Chimera window appear to save an image. Tom Jawahar Swaminathan wrote: > Dear Tom, > We are interested in making virus capsid pictures for the MSD Database > atlas pages using chimera. For this purpose, I was recently provided > with a script from Cathy Lawson (RCSB) which makes really pretty > images of viral capsids. I attach the script given by Cathy below. > > However, since we would want to run chimera in an automatic manner > (ideally with the --nogui) option, I've been trying to run the script > with --nogui and this appears to conflict with the module tkgui which > is called in the program. > > So my question: is it possible to make images without necessarily > opening up a GUI ? > > I would be really grateful if you could clear this us.. > > best regards, > > Jawahar Swaminathan > PDB Depositions > Macromolecular Structure Database > EBI

1 0

Problem with DCD not containing periodic box info when I use MD
by Antoine Daina 22 May '07

22 May '07

Hi Eric, I got a "DCD format error 9" returned by MD Movie in Chimera 1.2304 when trying to see the trajectory of a Charmm simulation without periodic box info. Can we easily fix it ? Antoine Daina, PhD, University of Geneva.

2 1

Tips for making density map images.
by Thomas Goddard 21 May '07

21 May '07

Here are some suggestions for making density map images in Chimera 1.2382. White background color. Set with two menu entries Actions/Color/Background and Actions/Color/White. To make white the default use Favorites/Preferences, category Background, set background color button to white and press Preferences dialog Save button. Surface smoothing. Use volume dialog Features/Surface-And-Mesh-Options and click "Surface smoothing..." checkbutton. To make this the default, turn it on then use Features/Save-default-dialog-settings. Close the surface-and-mesh-options panel before saving the settings or it will always be shown taking up alot of space. Image supersampling. Removes saw-tooth edges from saved images that are especially visible with white background. Does this by making a larger image in memory then averaging down to a smaller size. When saving an image you see many small pieces flash by in main window if supersampling is being used. To turn it on use File/Save-image, press the Image Setup button in the save image dialog, then set Supersample to 3x3 in the preferences dialog that appears. Press the Save button in the preferences dialog to make this the default. For Chimera versions after 1.2392 this will look different. Supersample will be an option on the save image dialog. Saving image same size as on screen. To make a saved image the same size in pixels as it appears in the Chimera window press the Get Pixels button in the File/Save-image dialog. In Chimera versions after 1.2392 this will be the default. Silhouette edges. To add black edges around maps that makes it easier to see when part of map is in front of another part use Favorites/Side-view, choose the Effects tab in the viewing dialog that appears, click on silhouette edges. Save raytraced image. Gives glossy appearance and shadows. Use File/Save-image and set "Raytrace with POV-Ray" to True. Make back parts of surface darker. Use Favorites/Side-view and move the right vertical yellow line towards the left. The surface gets dimmer from half way between the yellow lines to the right yellow line. Your eye position in this side view is shown as the square at the left and the red lines indicate how wide your field of view is. If you move the right yellow line (rear clipping plane) it will hide parts of the map -- everything to its right is invisible. Probably want to avoid that. Make recesses in surfaces darker to improve sense of depth. Use Tools/Volume-data/Surface-color. For spherical virus color the surface using "Radius", use 2 colors, Gray pattern, press "Set full range colormap values" to get minimum and maximum surface radius, press Color. You may need to press the Center button if the virus map is not centered at (0,0,0), and also press the "Set full range..." button again in that case. Probably you'll want to modify the minimum and maximum radius to get the best effect, also maybe adjust the two colors. For filaments use "Cylinder radius". Make recesses darker for asymmetric maps. This is a bit trickier than for spherical or cylindrical maps. For ribosomes or other asymmetric maps first create a blurred version of the map using Tools/Volume-data/Gaussian-filter. This tool has to be downloaded separately from the Chimera experimental features web page http://www.cgl.ucsf.edu/chimera/experimental/experimental.html and installed. Choose a gaussian width of about 1/3 the diameter of the surface (in angstroms, maybe 30 for a ribosome) and press Filter. Use the Tools/Volume-data/Surface-color and color the original map using the gaussian smoothed map. High values of the Gaussian smoothed map are in the interior and should be made a dark color. This makes the perception of depth dramatically better. Tom

1 0

Re: [Chimera-users] [chimera-dev] Protein to ligand shortest distances?
by Elaine Meng 21 May '07

21 May '07

Hi Kate, The Find Clashes/Contacts tool will report all atoms near your ligand and the corresponding distances. However, it does not work with a single strict cutoff, but instead uses the atoms' VDW radii. For example, if you used a cutoff of 0.0 overlap, that would find all atoms with VDW surfaces either intersecting or just touching. It finds all the atoms meeting your specified cutoff criteria, so you would still need to look at the list of results to see which atom is the closest from each residue. There are two cutoff parameters, the amount of VDW overlap and the extra closeness allowed for H-bonding pairs. As mentioned in the manual page, http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/findclash/ findclash.html the suggested values for these two parameters are: 0.4-1.0 and 0.2-0.6, respectively, if you only want to detect "clashes" (unfavorably close contacts) -0.4-1.0 (negative values) and 0.0, respectively, to detect contacts in general It sounds like you want the latter. Select your ligand (for example by Ctrl-click on one atom and then hitting the keyboard up arrow), open the tool (Tools... Structure Analysis... Find Clashes/Contacts), click "Designate" to specify you want to find contacts of the selected atoms (the ligand) versus all other atoms. Set the parameters to -0.5 and 0.0. In the bottom section check "Write information to reply log" and uncheck the other options. Open the reply log (it's in the Favorites menu). Click Apply and the list of results will be written in the Reply Log window. You could play around with the VDW overlap parameter and when you think you have what you want, you could use the "Write information to file" instead of the reply log. As I mentioned above, you would still have to look through the results to see the closest atom from each residue. The more negative the parameter, the more atoms will be found (including those which are further away from the ligand). There is also a command version of this tool, http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/findclash.html If you really care that you include all residues with any atom within 5 angstroms, you could do an additional step first: select the ligand, then Select... Zone using 5 angstroms, then use Actions... Write List to save the names of the "selected residues" to a file. You can compare this list to the more detailed results from Find Clashes/Contacts to find the closest atom from each of the residues. I hope this helps, Elaine p.s. I'm sending this to chimera-users too because that is more for user questions, while chimera-dev is more for developer/programmer questions ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html On May 20, 2007, at 10:12 PM, khar6039(a)mail.usyd.edu.au wrote: > Hi all, > I'm new to chimera and would like a simple way to zone 5 Angstroms > around a > ligand and then have chimera tell me what the shortest distances are > between these residues and the ligand. Is this possible? > Cheers, > Kate

1 0

Multiple stucture comparision
by Pankaj Barah 19 May '07

19 May '07

Dear all, I have been using UCSF Chimera for structural superpositions for last few days. But while trying to do multiple superimposition for a set of proteins, i am unable to perform it. can anybody give me any suggestion? Regards Pankaj Barah Junior Research Fellow, Phone:+91-9922346120(M) Bioinformatics Centre, +91-20-25692039(O) University of Pune, Pune 411 007,Maharashtra, India. Pankaj Borah MSc Bioinformatics Univ of Madras(HCAS) Mobile number: 09884270481(Chennai)/ 09844514157(B'lore) Home number:(03752)259201 Alt. Mail id: pankajborah(a)gmail.com Group homepage: http://groups.yahoo.com/group/bioinformatrix/ --------------------------------- Office firewalls, cyber cafes, college labs, don't allow you to download CHAT? Here's a solution!

2 1

Pub3D database URL
by Elaine Meng 18 May '07

18 May '07

Hi everybody, In a previous message I mentioned Pub3D as a useful source of 3D coordinates (SDF files) for many PubChem entries. The URL for that server is now http://www.chembiogrid.org/cheminfo/p3d Enjoy! Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html Begin forwarded message: > From: Rajarshi Guha <rguha(a)indiana.edu> > Date: May 17, 2007 4:44:53 PM PDT > To: Elaine Meng <meng(a)cgl.ucsf.edu> > Subject: Re: Pub3D SDFs > >> I noticed that your Pub3D site has not been reachable for a while. >> ( http://rguha.ath.cx/~rguha/p3d/ ) > > Aah! My Apache redirect failed again :( > > Actually, we have a new URL for it which should be constant (and is > independent of my machine) > > http://www.chembiogrid.org/cheminfo/p3d > > ------------------------------------------------------------------- > Rajarshi Guha <rguha(a)indiana.edu> > GPG Fingerprint: 0CCA 8EE2 2EEB 25E2 AB04 06F7 1BB9 E634 9B87 56EE > -------------------------------------------------------------------

2 2