- Chimera-users - RBVI Mailing Lists

Action focus
by Thomas Goddard 15 May '08

15 May '08

I made Action / Focus work on selected surfaces. Also I made the none-implies-all rule include surfaces in the "all". So Actions / Color / green will color everything green including surfaces if nothing is selected. I fear this a bit because volume user often have tens of maps and pdb models open and could shoot themselves in the foot pretty easily. These changes were a bit more complicated than expected and may be a buggy. Report anything suspicious or wait until next week when I shake it down a bit more. Tom

1 0

electrostatic potential colored surfaces script
by Miguel Ortiz-Lombardía 15 May '08

15 May '08

Dear all, I wonder if someone has written and would share a script to sequentially read a pdb file, show its molecular surface and color it by the electrostatic potential from an appropriate .dx file. Cheers, Miguel -- http://www.pangea.org/mol/spip.php?rubrique2 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Je suis de la mauvaise herbe, Braves gens, braves gens, Je pousse en liberté Dans les jardins mal fréquentés! Georges Brassens

2 2

Help needed
by Bagchi, Angshuman 13 May '08

13 May '08

Hi, I am a student of IUPUI (indiana-University-Purdue University-Indianapolis). I would like to know how chimera can be used to select the amino acid residues of one chain of a protein, which are within a certain distance (say 10A) of another chain of the same protein. For example, the pdb 1o6s has two chains A and B. I would like to find what amino acid residues of A chain of 1o6s are within 10A of B chain of 1o6s. I am eagerly expecting a reply from your end. Thanks Ang

3 2

Drawing a circle or square
by Elaine Meng 12 May '08

12 May '08

Hi Pradeep, Other people have recommended Adobe Illustrator and Inkscape for such purposes: http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-April/002579.html However, you CAN draw a rectangle in the BILD format and read that into Chimera. For example, put the following 7 lines in a text file named square.bild and open it with File... Open: .scale 1.5 .color red .move 0 0 0 .draw 5 0 0 .draw 5 5 0 .draw 0 5 0 .draw 0 0 0 I also attach said file below, for your convenience. You can edit the coordinates, scale, and color. Once you open it in Chimera, it is a separate model, and you can move it independently using the checkboxes below the Command Line or the "Active" checkboxes in the Model Panel. Advantage: since it does exist in 3D, you can place it between the surface and the nucleic acid. Disadvantage: cannot make curves/circles, may be hard to position/orient, cannot control line thickness. The thickness issue may be addressed by instead making the edges cylinders and the corners balls. This can also be done in the simple BILD format, but it is sort of a hack to make a 3D object when you really want a 2D object. BILD format description: http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/bild.html I hope this helps, 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 On Apr 30, 2008, at 11:57 AM, Pallan, Pradeep S wrote: > Hi Elaine, > I am wondering whether one can draw a circle or a square > to indicate a site/s of a protein? Attached is a tiff file that has a > square and a circle. I want the regions to be pointed out. > > So what I did: > 1) Generated the image in chimera > 2) added a circle and a square in microsoft paint (Doesn't look > good!). > > I was going through the chimera tutorials, but didn't find the > feature. > Any suggestions for adding a circle or square? In my case, the > surface image is at the back, and nucleic acid in the front. > I would like to draw the circle and square and then load the nucleic > acids, so that will look better. In other words the square or > circle in > between > the surface image (at the back) and the nucleic acid (in front). > > If editing the tiff file using Adobe photoshop (instead of MS paint) > it does a better job, thick lines etc, but it leaves patches on the > nucleic > acid. > I appreciate your help. > Thanks, > Pradeep

2 1

measure volume area request
by Tom Goddard 12 May '08

12 May '08

Edward Brignole wrote: > Tom, > > I have a request for display of the measure volume area tool - can it > be moved from "tools" to "features" such that it is displayed in the > volume viewer along with color, voxel size, etc. The problem I have > is that frequently I have 30 or so volumes to open at once and then > need to independently go through them and set density thresholds. > It's really a pain to have to select the volume once in the volume > viewer and then again select the same volume in the measure tool then > go back to the viewer and adjust threshold. After doing this 30 > times, you'll find that most of the effort is spent making sure to > select the same volume in both menus so that the selected volume that > you are adjusting is the same volume being measured. > > Thankfully, I did however find a python script that you emailed a year > ago that I can steal a few lines out of to adjust the thresholds > automatically so that the enclosed surface of each model has the > appropriate volume in A^3. BTW, I really like your hidedust.py as > well, however once I've executed that script adjusting the threshold > in the volume viewer does nothing to the displayed or measured volumes > - I'm using the snapshot from a week or so ago. > > Ed > > Ed Brignole > Asturias Lab, CB227 > The Scripps Research Institute > Hi Ed, I agree it would be helpful to have a measured volume/area panel in the volume dialog. I'll add that to our requested features list. http://www.cgl.ucsf.edu/chimera/plans.html It still needs to be a separate dialog too because it works on surfaces that are not from volume data (e.g. molecular surfaces). The hidedust.py script hides connected components smaller than a specified volume. But the Measure Volume and Area dialog displays values for the entire surface, hidden or not. So those values don't change. Currently there is no provision in the volume/area measurement to exclude the hidden part of the surface. You also say that changing the contour level does not change the displayed surface. As soon as you change the contour level the whole surface is replaced. The hidedust.py script only acts on the surface that is shown at the time you run it, not on future surfaces, so all the dust should come back. You seem to say this is not what happens in your use. I've attached the hidedust.py script to this email in case you are using a different version. Tom # ----------------------------------------------------------------------------- # Hide connected components of a surface with small enclosed volume. # # Tested with Chimera version 1.2500. Will not work with 1.2470 or earlier. # def hide_small_blobs(p, volume): varray, tarray = p.geometry from numpy import ones, intc mask = ones((len(varray),), intc) from _surface import connected_pieces cplist = connected_pieces(tarray) hid = 0 for vi, ti in cplist: ta = tarray.take(ti, axis = 0) from MeasureVolume import enclosed_volume vol, holes = enclosed_volume(varray, ta) if not vol is None and vol < volume: mask.put(vi, 0) hid += 1 p.setTriangleMaskFromVertexMask(mask) print ('Hid %d of %d connected surface components having volume < %.5g' % (hid, len(cplist), volume)) # ----------------------------------------------------------------------------- # Return all surface pieces if none are selected. # def selected_surface_pieces(): import Surface plist = Surface.selected_surface_pieces() if len(plist) > 0: return plist from chimera import openModels from _surface import SurfaceModel plist = [] for m in openModels.list(modelTypes = [SurfaceModel]): plist.extend(m.surfacePieces) return plist # ----------------------------------------------------------------------------- # volume = 100.0 for p in selected_surface_pieces(): hide_small_blobs(p, volume)

1 0

Encoding movie by hand
by Tom Goddard 12 May '08

12 May '08

Anindito Sen wrote: > > Hi Tom > > How can we make a movie (not morphing) out of several still snaps > (.jpg) saved from of many density maps. > using chimera. > > Thanks > qandy > > Dr. Anindito Sen (Ph.D) > Research Associate , Dept. of Biochemistry and Molecular Genetics > University of Virginia > Box 800733 Charlottesville, VA 22908 Hi Qandy, You can create a movie directly from a set of images using the ffmpeg executable included in the Chimera distribution. When you make a movie using Chimera it runs the ffmpeg executable to make the movie file and the options used are printed to the Chimera reply log (Favorites / Reply Log). For example, in the reply log I see ARG_LIST IS ['-r', '25', '-i', '/tmp/chimovie_JgKl-%05d.ppm', '-y', '-f', 'mov', '-b', '2000', '-bufsize', '200', '/Users/goddard/chimera_movie.mov'] Movie saved to /Users/goddard/chimera_movie.mov so the ffmpeg command used was: ffmpeg -r 25 -i /tmp/chimovie_JgK1-%05d.ppm -y -f mov -b 2000 -bufsize 200 You could run a similar command from a shell to encode a movie outside of Chimera. The files are specified by the "-i" option. In the example they are ppm format but jpg will work too. The %05d means that the file names ended in a 5-digit number with leading zeros (e.g. chimovie_JgK1-00000.ppm, chimovie_JgK1-00001.ppm, ...). The -r option indicates frame rate (25 frames per second). The -f option is output format (mov = quicktime). The -b option is bit rate (Kbits/sec) and controls output quality. The -y option means to overwrite the output file if it already exists. For complete ffmpeg options see online docs: http://ffmpeg.mplayerhq.hu/ffmpeg-doc.html The Chimera copy of ffmpeg is in the distribution in chimera/bin (or on Mac in Chimera.app/Contents/Resources/bin). Tom

1 0

Program won't open
by Abigail Lambert 12 May '08

12 May '08

Hello - I'm using a MacBook Pro, OS X (10.4.11), and I had been successfully using Chimera for a while. Then all of sudden, the program will no longer open. The Chimera icon blinks once in the taskbar, and then it disappears and the program does NOT open. I tried uninstalling and re-installing, but this does not help. Any suggestions? Thanks! - Abbie -

2 1

Renaming chains in a PDB file with two chains
by Harinathachari Bahudhanapati 10 May '08

10 May '08

Hi Elaine, Thanks for all the answers you give to our questions. They are very helpful as I learn more everyday. I have a question. Is there anyway to rename a chain in a PDB file through the GUI? I have a few PATCHDOCK generated PDB files with two different chains named under Chain 'A' as #0.1 and #0.2. When I try to get the list of hydrogen bonds between these two chains, the chain name 'A' appears for all of the h-bonding pairs. So, I want to have them named Chain A and Chain B for easier analysis than scratching my head figuring out which h-bond pair is between the chains or with in the chain. I am attaching the pdb file. Please help me. (I thought of renaming the chains in the PDB file going through the PDB header (text file) and manually changing the name. But I am not a expert if anything goes wrong.) Sincerely Hari

2 2

Dashed H-bond
by Ibrahim Moustafa 09 May '08

09 May '08

Dear Chimera Team, Few weeks ago I asked about how to make H-bonds with dashed line style, I got your great help in that. So, I can make the H-bond with the color and style I want; however, when saving the image (with POV true) I got the H-bond appeared with the new chosen color but not as dashed line shown in the display. But, if I save the image with POVray option false, I got the same style shown in the display (new color and dashed line). The question is why saving the image using ³POV true² does not give the image as it appears in the display? Again, it accepts the new color but not the dashed style. Thanks for your great help, Ibrahim -- Ibrahim M. Moustafa, Ph.D. Biochemistry and Molecular Biology Dept. 201 Althouse Lab., University Park, Pennsylvania State University PA 16802 Tel. (814) 863-8703 Fax (814) 865-7927

3 3

Number of voxels?
by Kolodziej, Steven J 09 May '08

09 May '08

I was wondering if anyone knows how to tell how many voxels within a bounding box can be assigned to protein signal instead of "empty space" at any given contour/density level? Thanks, Steve Kolodziej

2 1