- Chimera-users - RBVI Mailing Lists

interactive volume segmentation
by Juha Huiskonen 18 Dec '03

18 Dec '03

Hello, currently it is possible to select only box-shaped subregions in Volume Viewer. I would be especially interested in segmenting electron density maps or their isosurfaces to different subregions with any shape interactively. Is anyone planning to add a feature like this? Regards, Juha Huiskonen

2 3

Re: [Chimera-users] Chimera export to VRML?
by Thomas Goddard 16 Dec '03

16 Dec '03

Hi Morgan, I've written an extension to export Chimera multi-scale and volume surfaces as VRML. I wasn't able to export the molecular (MSMS) surfaces because the code does not provide access to the surface coloring. I've submitted a request to enhance the molecular surface code to make surface colors readable. I'm not sure if the VRML surface exporting will make it into the coming snapshot release. If it doesn't I will provide it to you by email. I'll let you know when we have the snapshot ready, probably later this week. Tom

1 0

Re: [Chimera-users] Chimera export to VRML?
by Thomas Goddard 16 Dec '03

16 Dec '03

Hi Morgan, We hope to have a snapshot Chimera release later this week. I'll try to get VRML output of multi-scale and msms surfaces in that release. Snapshot releases aren't as well tested as production releases, but are intended for trying out new features like this. Tom PS: Your email does not go to chimera-users unless you explicitly CC it to that list. We like to post to that list so others can benefit from it. I've included your email below and CC'ed this to chimera-users. Let me know if you want to keep your email private. > From: Morgan Ryan <mryan1(a)nc.rr.com> > Subject: Re: [Chimera-users] Chimera export to VRML? > Date: Tue, 16 Dec 2003 14:35:27 -0500 > > Thanks, Thomas. (Am I broadcasting to the list? I'm not sure if I'm > talking to a list or just you and Eric Petterson...) > > > For a large structure like a > > ribosome the VRML file ...might have a million triangles, and the file > > will > > at least be 30 megabytes. > > The 3D programs I use have fairly refined mesh management tools for > reducing complexity selectively, although being able to choose the mesh > density directly through MSMS would be pleasing. > > > You can choose the resolution of the multi-scale surfaces. We don't > > currently have VRML output for the multi-scale surfaces but that > > could be added fairly easily. I could look into that. > > The advantage of VRML output (or 3dmf, or dxf, or some other > interchange format) is that you could open the model in Maya, Studio > Max, Strata, Cinema 4D, any of the 3D animation programs, and make > startling movies wiith Hollywood special effects--light sabers! > Atomic-resolution recreations of enzyme reaction mechanisms! (The > latter is what I'm working on...) > > It would be a wonderful extension of Chimera if VRML output were added. > Morgan

1 0

Chimera export to VRML?
by Morgan Ryan 16 Dec '03

16 Dec '03

VRML seems to be a native internal format for (the magnificent) Chimera. Is it possible to export VRML models from Chimera? Or do you worldly folks know of a mechanism for converting msms output into vrml files? Thanks! Morgan Ryan, enthusiastic amateur ____________________ Morgan Ryan Development editor W H Freeman 209 Adams Way Chapel Hill NC 27516 phone: 919 942 4890 fax: 651 846 5265 mryan(a)whfreeman.com

3 3

help
by amutha ramaswamy 16 Dec '03

16 Dec '03

Dear sir I am using the Chimera molecular visualising software. I am not able to color the PDB structure according to the B-factors. could you plase let me know how do i color accroding to the bfactor. i would greatly apprecuate your help in this regard sincerely amutha Dr Amutha Ramaswamy Postdoctoral Fellow, Department of Biomedical Informatics 3184 Graves Hall, 333 West 10th Avenue The Ohio state University, Columbus, OHIO 43210 Phone: 614-688-3236 (Office); 614-291-0071(Residence) Fax: 614-292-7659 (Office) --------------------------------- Do you Yahoo!? New Yahoo! Photos - easier uploading and sharing

2 2

Re: Gaussians
by Thomas Goddard 10 Dec '03

10 Dec '03

Hi Lars, I'm not sure what you mean by visualizing some Gaussians. If you mean you want to make a volume which contains some specified Gaussians then I can give you a separate Python program that will do that. I already wrote it for another user. It is included at the end of this email. It outputs a netcdf format volume file. Tom % ./spots.py Make a volume array that is the sum of specified Gaussians. The first line of the input file has the grid cell size. Each subsequent line has 5 fields giving x, y, z, width, height. A volume grid is made that includes each Gaussian out to 3 widths. The resulting volume is written in netcdf format. Syntax: spots.py <infile> <outfile> % cat spotfile .5 15.49100 7.05000 17.16500 0.38 1.00 14.94600 7.55000 15.88800 0.35 1.00 16.01100 7.60400 14.81400 0.34 1.00 ... % ./spots.py spotfile gspottest.nc The spots.py code follows. Note that it contains a path to your Chimera installation because it needs Chimera modules to write the netcdf volume. #!/usr/bin/env python # ----------------------------------------------------------------------------- # Make a volume array that is the sum of specified Gaussians. # The resulting volume is written in netcdf format. # # Syntax: spots.py <infile> <outfile> # # The input file has a grid cell size on the first line followed by one # line per gaussian with 5 fields specifying x, y, z, width, height. # import os.path import sys # Need Chimera path to use VolumeData and Numeric modules CHIMERA_PATH = '/usr/local/chimera' sys.path.insert(0, os.path.join(CHIMERA_PATH, 'lib')) sys.path.insert(0, os.path.join(CHIMERA_PATH, 'share')) import Numeric # ----------------------------------------------------------------------------- # def make_spot_grid(path): step_size, spot_parameters = read_spots(path) range = 3 # in spot widths xyz_min, xyz_max = bounding_box(spot_parameters, range) xyz_origin = xyz_min xyz_step = (step_size, step_size, step_size) xyz_size = map(lambda a,b: a - b, xyz_max, xyz_min) import math grid_size = map(lambda s, step, math=math: 1 + int(math.ceil(s/step)), xyz_size, xyz_step) a = spot_array(spot_parameters, range, xyz_origin, step_size, grid_size) from VolumeData import Array_Grid_Data g = Array_Grid_Data(a, xyz_origin, xyz_step) return g # ----------------------------------------------------------------------------- # def read_spots(path): f = open(path, 'r') lines = f.readlines() f.close() step_size = float(lines[0]) spot_params = [] for line in lines[1:]: params = map(float, line.split()) spot_params.append(params) return step_size, spot_params # ----------------------------------------------------------------------------- # def bounding_box(spot_parameters, range): xmin = ymin = zmin = None xmax = ymax = zmax = None for x, y, z, w, h in spot_parameters: pad = range * w if xmin == None or x - pad < xmin: xmin = x - pad if ymin == None or y - pad < ymin: ymin = y - pad if zmin == None or z - pad < zmin: zmin = z - pad if xmax == None or x + pad > xmax: xmax = x + pad if ymax == None or y + pad > ymax: ymax = y + pad if zmax == None or z + pad > zmax: zmax = z + pad return (xmin, ymin, zmin), (xmax, ymax, zmax) # ----------------------------------------------------------------------------- # def spot_array(spot_parameters, range, xyz_origin, step_size, grid_size): shape = list(grid_size) shape.reverse() # array indices are z, y, x. array = Numeric.zeros(shape, Numeric.Float) for x, y, z, width, height in spot_parameters: add_spot((x,y,z), width, height, range, xyz_origin, step_size, array) return array # ----------------------------------------------------------------------------- # def add_spot(xyz, width, height, srange, xyz_origin, step_size, array): ijk = map(lambda a,b,s=step_size: (a - b)/s, xyz, xyz_origin) w = width / step_size w2 = w * w r = srange * width / step_size import math imin, jmin, kmin = map(lambda l, r=r, math=math: int(math.floor(l-r)), ijk) imax, jmax, kmax = map(lambda l, r=r, math=math: int(math.ceil(l+r)), ijk) for k in range(kmin, kmax+1): for j in range(jmin, jmax+1): for i in range(imin, imax+1): di, dj, dk = i-ijk[0], j-ijk[1], k-ijk[2] s2 = di*di + dj*dj + dk*dk e = height * math.exp(-s2/(2*w2)) array[k,j,i] += e # ----------------------------------------------------------------------------- # def bfactor_backbone_spots(path, molecule = None): if molecule == None: import chimera molecule = chimera.openModels.list()[0] out = open(path, 'w') out.write('.5\n') # step size import math for a in molecule.atoms: if a.name in ('C', 'N', 'CA') and hasattr(a, 'bfactor'): x, y, z = a.coord().xyz.data() w = math.sqrt(a.bfactor / (8 * math.pi**2)) h = 1 out.write('%8.5f %8.5f %8.5f %6.2f %6.2f\n' % (x,y,z,w,h)) out.close() # ----------------------------------------------------------------------------- # def syntax(): msg = ('Make a volume array that is the sum of specified Gaussians.\n' + 'The first line of the input file has the grid cell size.\n' + 'Each subsequent line has 5 fields giving x, y, z, width, height.\n' + 'A volume grid is made that includes each Gaussian out to 3 widths.\n' + 'The resulting volume is written in netcdf format.\n' + 'Syntax: spots.py <infile> <outfile>\n') sys.stderr.write(msg) sys.exit(1) # ----------------------------------------------------------------------------- # def run_command(): if len(sys.argv) != 3: syntax() inpath = sys.argv[1] outpath = sys.argv[2] grid = make_spot_grid(inpath) from VolumeData.netcdf.netcdf_grid import write_grid_as_netcdf write_grid_as_netcdf(grid, outpath) # ----------------------------------------------------------------------------- # if __name__ == '__main__': run_command()

1 0

B-Factor
by Luis Cuello 08 Dec '03

08 Dec '03

Hello, My name is Luis Cuello and I am a graduate student at the University of Virginia..I started using chimera recently but I have not been able to colored my files by B-factors values, Is there any way to do that with chimera software??? If you could help me in this matter I would appreciated..Sincerely Luis Cuello University of Virginia Molecular Physiology 1(434)9245473 e-mail:lgc4r@virginia.edu

2 1

Re: [Chimera-users] delphi potenital map viewing
by Thomas Goddard 05 Dec '03

05 Dec '03

Hi Klaas, The Delphi Viewer is the only tool that can color a molecular surface according to the potential value. Can you email me the Python traceback you get when the "_cubes error: ..." occurs? This shows exactly where in the code the error happened. You find it by looking at Favorites/Reply Log. Having the traceback alone might be enough for me to figure out and correct the problem. If you could put the phi file on the web so I could download it I would almost certainly be able to figure out the problem. The Delphi Viewer is supposed to be able handle maps of any size. It will be much slower than volume viewer, but it should not generate an error with a large map. Tom > Date: Fri, 05 Dec 2003 21:10:20 +0100 > From: Klaas Decanniere <Klaas.Decanniere(a)pandora.be> > Reply-To: Klaas.Decanniere(a)pandora.be > To: Thomas Goddard <goddard(a)cgl.ucsf.edu> > Subject: Re: [Chimera-users] delphi potenital map viewing > > Thomas Goddard wrote: > > >Hi Klaas, > > > > The size of the Delphi potential map Chimera can display is only > >limited by the size of your computer memory. > > > > If you open a large map it does not get shown immediately if the map > >is larger than 1 Mvoxels (~100^3). > > > > > > When I tried using the "delphi viewer" option, I get _cubes error: grid > index out of range > Using the volume viewer I can indeed load everything. > Is there a way to tie the volume viewer thing to a molecular surface? > With a colour gradient showing different levels of potential? (As in the > Molecular surface option in the Delphiviewer) > Or maybe I can save the subsampled map somehow and then read it into the > delphiviewer thing? > > Thanks again for your help. > > Klaas > > P.S.: I like the program more each day. > P.P.S.: If you ever have a day with nothing to do: the automatic > fetching of pdb files doesn't work for me as I'm behind a proxy server.

1 0

delphi potenital map viewing
by Klaas Decanniere 04 Dec '03

04 Dec '03

Hello, Thinking that more is better I ended up with Delphi potential maps on a rather large grid which I would like to visualize with chimera (gsize=213) Chimera seems to think the grid is to large. Is there an easy way to have chimera accept bigger grids (speed is not really an issue)? Alternatively, what is the largest grid that chimera can accept? Thanks for your help. Klaas Decanniere

2 1

ribbon+atoms
by Elaine Meng 25 Nov '03

25 Nov '03

Dear Soren, Thanks for the compliments - we're trying! You're right that in recent versions of Chimera , showing ribbon turns off the backbone atoms. There is an easy workaround: open two copies of the same molecule, use one copy to show ribbon and the other copy to show atoms. The reason for the behavior is when the ribbon is drawn as a smooth curve, it does not exactly pass through the backbone atoms. Often one wants to show just the ribbon and certain side chains, and the current behavior is to extend the side chain to meet the ribbon (even when the backbone atoms are not on the ribbon) so that they are not floating in space. However, we agree that sometimes one will want to display both the ribbon and the backbone atoms, and in the future there will a preference to switch between: (a) show ribbon, not backbone atoms, and extend the side chains to meet the ribbon (b) show ribbon and whatever backbone atoms are displayed by the user, do not extend side chains to meet the ribbon Until this switchable preference becomes available, we hope that the suggested workaround will let you do what you want. Sincerely, Elaine =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Elaine C. Meng, Ph.D. Babbitt Lab and Computer Graphics Lab meng(a)cgl.ucsf.edu http://www.cgl.ucsf.edu/home/meng/index.html =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

1 0