Hi, Chimera and PyMol are showing very different thermal ellipsoid orientations for the same PDB file, and I have reason to believe that the PyMol display is the right one. Here are the first two lines of the PDB file I am reading in: Atom 1 O1 CUC 1 3.345 1.613 0.172 ANISOU 1 O1 CUC 1 619 156 528 150 43 -6 ATOM 2 O2 CUC 1 -1.699 0.125 2.792 ANISOU 2 O2 CUC 1 497 355 848 128 -68 204 I would appreciate help with this because Chimera appears to provide more detailed control of the coloring of the ellipsoids than PyMol does. I'll be happy to provide screenshots or other information. Regards, Mike
Hi Mike, After comparing with ellipsoids from various other programs I now do see some differences. I've open a Trac ticket (#7446) for this problem so that you will be notified of any progress on the issue. I should have some kind of resolution in the next day or three. If you wanted to send along some examples of expected output I would be happy to check any fix I make against them. --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 5, 2009, at 11:37 PM, Michael K. Gilson wrote:
Hi,
Chimera and PyMol are showing very different thermal ellipsoid orientations for the same PDB file, and I have reason to believe that the PyMol display is the right one. Here are the first two lines of the PDB file I am reading in:
Atom 1 O1 CUC 1 3.345 1.613 0.172 ANISOU 1 O1 CUC 1 619 156 528 150 43 -6 ATOM 2 O2 CUC 1 -1.699 0.125 2.792 ANISOU 2 O2 CUC 1 497 355 848 128 -68 204
I would appreciate help with this because Chimera appears to provide more detailed control of the coloring of the ellipsoids than PyMol does.
I'll be happy to provide screenshots or other information.
Regards, Mike
_______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
It turns out that the math that transformed a uniform sphere into an ellipsoid was over-complicated and wrong. That'll teach me to crib code from other programmers in the lab. ;-) Also, I wasn't taking the square root of the eigenvalues like I should have, since the U matrix is in units of length squared. I committed fixes for both problems and if tonight's build works then they'll be in the daily build available tomorrow. Check that the date in the table is at least July 6th. Otherwise the build failed and you'd have to try again the next day. --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 5, 2009, at 11:37 PM, Michael K. Gilson wrote:
Hi,
Chimera and PyMol are showing very different thermal ellipsoid orientations for the same PDB file, and I have reason to believe that the PyMol display is the right one. Here are the first two lines of the PDB file I am reading in:
Atom 1 O1 CUC 1 3.345 1.613 0.172 ANISOU 1 O1 CUC 1 619 156 528 150 43 -6 ATOM 2 O2 CUC 1 -1.699 0.125 2.792 ANISOU 2 O2 CUC 1 497 355 848 128 -68 204
I would appreciate help with this because Chimera appears to provide more detailed control of the coloring of the ellipsoids than PyMol does.
I'll be happy to provide screenshots or other information.
Regards, Mike
_______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hello! I did not know Chimera was able to display thermal ellipsoids at all. How can I do so? Is there some kind of tutorial available for this? Will the ellipsoid information be read from a cif file too, or does it have to be a pdb file? Kind regards, Thomas
Hi Thomas, It is a new feature, so only in the daily builds (and only corrected as of July 6, so the very latest build): <http://www.cgl.ucsf.edu/chimera/alpha-downloads.html> Currently it works only using the information in ANISOU records of PDB files. You would just open the PDB file as usual, then use the Thermal Ellipsoids tool (under Tools... Structure Analysis). It is fairly self-explanatory, but there is also documentation -- just click the Help button on the dialog or see the copy at our site: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/thermal/thermal.htm...
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html On Jul 7, 2009, at 3:12 AM, Dipl.-Ing. Thomas Mitterfellner wrote:
Hello! I did not know Chimera was able to display thermal ellipsoids at all. How can I do so? Is there some kind of tutorial available for this? Will the ellipsoid information be read from a cif file too, or does it have to be a pdb file? Kind regards, Thomas
Hello again!
Currently it works only using the information in ANISOU records of PDB files. You would just open the PDB file as usual, then use the Thermal Ellipsoids tool (under Tools... Structure Analysis). It is fairly self-explanatory, but there is also documentation -- just click the Help button on the dialog or see the copy at our site: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/thermal/thermal.htm...>
I'm very happy to see this working! At the risk of sounding ungrateful -- I'd like to request two features. The first one: it is quite common to plot thermal ellipsoids on some probability level (e.g. 30% or 50%). Unfortunately, I don't know how these Values correspond to the scaling factor that currently controls the size of the ellipsoids. Could you implement this kind of scaling as a feature or point out, if/how these two measures correspond to each other? The second one: could you possibly implement it as a feature, that the lines resulting from the intersection of the main planes (xy, xz, yz) with the surface of the ellipsoid (the principal ellipses) can be displayed ? Thank you very much, kind regards, Thomas
On Jul 7, 2009, at 10:09 AM, Dipl.-Ing. Thomas Mitterfellner wrote:
Hello again!
Currently it works only using the information in ANISOU records of PDB files. You would just open the PDB file as usual, then use the Thermal Ellipsoids tool (under Tools... Structure Analysis). It is fairly self-explanatory, but there is also documentation -- just click the Help button on the dialog or see the copy at our site: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/thermal/thermal.htm...
I'm very happy to see this working!
At the risk of sounding ungrateful -- I'd like to request two features.
The first one: it is quite common to plot thermal ellipsoids on some probability level (e.g. 30% or 50%). Unfortunately, I don't know how these Values correspond to the scaling factor that currently controls the size of the ellipsoids. Could you implement this kind of scaling as a feature or point out, if/how these two measures correspond to each other?
We intend to implement a probability-to-scaling-factor calculator in the interface, but haven't gotten there yet. We felt that the tool was useful enough as is to put it out in what we consider to be an incomplete state. As a stopgap for you, the scaling factor for 30% is 1.1932 and for 50% is 1.5382.
The second one: could you possibly implement it as a feature, that the lines resulting from the intersection of the main planes (xy, xz, yz) with the surface of the ellipsoid (the principal ellipses) can be displayed ?
Given the way the underlying ellipsoid depiction is generated, it won't be easy. Nonetheless, I do think it should happen but it will take awhile. One thing that would be easy for me to do would be to actually show the ellipsoid axes (as lines). Would this help any if I put it in there? --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
Hello! Thanks for the information concerninge the scaling factors.
Given the way the underlying ellipsoid depiction is generated, it won't be easy. Nonetheless, I do think it should happen but it will take awhile. One thing that would be easy for me to do would be to actually show the ellipsoid axes (as lines). Would this help any if I put it in there?
If I understand it correctly, the ellipsoid axes you mentioned would be invisible if I the ellipsoids are rendered 100% opaque, right? So it for me, personally, it won't help, but probably for others. I was just thinking about the three ellipses that make up each ellipsoid to be shown on the surface. Concerning the CIF file, I was talking about a small molecule CIF. Unfortunately, I don't know about the mmCIF format. The information in my CIF is encoded like this: <snip> _atom_site_label _atom_site_type_symbol _atom_site_fract_x _atom_site_fract_y _atom_site_fract_z _atom_site_U_iso_or_equiv _atom_site_adp_type _atom_site_occupancy _atom_site_symmetry_multiplicity _atom_site_calc_flag _atom_site_refinement_flags _atom_site_disorder_assembly _atom_site_disorder_group Si1 Si 0.27946(5) 0.08326(9) 0.39479(4) 0.0163(3) Uani 1 1 d . . . H1 H 0.2704 0.1083 0.4766 0.034 Uiso 1 1 calc R . . </snip> This is an example for one anisotropic atom and an isotropic one. Later on in the file, from how I understand it, there seems to be the actual ellipsoid information: <snip> _atom_site_aniso_label _atom_site_aniso_U_11 _atom_site_aniso_U_22 _atom_site_aniso_U_33 _atom_site_aniso_U_23 _atom_site_aniso_U_13 _atom_site_aniso_U_12 Si1 0.0181(5) 0.0176(5) 0.0126(5) -0.0010(3) 0.0038(4) -0.0003(4) </snip> Kind regards, Thomas
Eric and Elaine, The display of axes and major ellipses as well as a probability-level- to-scaling-factor calculator will be a great addition to Chimera and I think more small molecule people and Organic chemists will start to use the program. I'm really excited about this feature being added to Chimera and I wanted to pass along some ideas for displaying the axes as well as octants. The program RASTEP (Raster3D Thermal Ellipsoid Program) has some very nice rendering features that would be great options to have in chimera. I attached an example below. Thanks to the Chimera team for working on this feature, Dave Chenoweth
Hi David, I guess my question is how important is fancy5 and fancy6? Those will be the hard ones for me to do. Keep in mind that "display of ... major ellipses" will look like fancy3 except that you will be able to see the arcs of the major ellipses without using a transparent ellipsoid. The other "fancy modes" are not difficult to do. --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 8, 2009, at 8:29 AM, David Chenoweth wrote:
Eric and Elaine,
The display of axes and major ellipses as well as a probability- level-to-scaling-factor calculator will be a great addition to Chimera and I think more small molecule people and Organic chemists will start to use the program. I'm really excited about this feature being added to Chimera and I wanted to pass along some ideas for displaying the axes as well as octants. The program RASTEP (Raster3D Thermal Ellipsoid Program) has some very nice rendering features that would be great options to have in chimera. I attached an example below.
Thanks to the Chimera team for working on this feature, Dave Chenoweth
Eric, For small molecule/high resolution work an ortep type representation would be as shown below (all three clips are from the same molecule). The ellipsoid, it's three axis and a shaded octant. I understand the octant is the hardest part. As you can see the addition of thermal ellipsoid information adds a lot to the rendering. My small molecule users are very excited about the addition of thermal ellipsoids to Chimera and we're doing hundreds of structures per year. Cheers, Mike <<< ------------------------------------------------------------------------>
Dr. Michael W. Day Director - X-ray Crystallography Lab & Molecular Observatory California Institute of Technology Mail Code 139-74 Pasadena, CA 91125
<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< Beckman Institute, Room 116 Phone: (626) 395-2734 Fax: (626) 449-4159 e-mail: mikeday@caltech.edu <<< ------------------------------------------------------------------------>
On Jul 8, 2009, at 9:56 AM, Eric Pettersen wrote: Hi David, I guess my question is how important is fancy5 and fancy6? Those will be the hard ones for me to do. Keep in mind that "display of ... major ellipses" will look like fancy3 except that you will be able to see the arcs of the major ellipses without using a transparent ellipsoid. The other "fancy modes" are not difficult to do. --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 8, 2009, at 8:29 AM, David Chenoweth wrote:
Eric and Elaine,
The display of axes and major ellipses as well as a probability- level-to-scaling-factor calculator will be a great addition to Chimera and I think more small molecule people and Organic chemists will start to use the program. I'm really excited about this feature being added to Chimera and I wanted to pass along some ideas for displaying the axes as well as octants. The program RASTEP (Raster3D Thermal Ellipsoid Program) has some very nice rendering features that would be great options to have in chimera. I attached an example below.
Thanks to the Chimera team for working on this feature, Dave Chenoweth
<Thermal_ellipsoid_representations.tiff> _______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
On Jul 7, 2009, at 3:12 AM, Dipl.-Ing. Thomas Mitterfellner wrote:
Will the ellipsoid information be read from a cif file too, or does it have to be a pdb file?
As Elaine mentioned, PDB file right now. We probably should get our mmCIF support to handle the anisotropic information, which is isn't yet. So are you talking about a small-molecule CIF file or an mmCIF file? If CIF, do you know if it encodes the anisotropic information in the same way as mmCIF? --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
participants (6)
-
David Chenoweth -
Dipl.-Ing. Thomas Mitterfellner -
Elaine Meng -
Eric Pettersen -
Michael Day -
Michael K. Gilson