How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid. While I'm on the ellipsoid topic; has there been progress on overlaying axis on the ellipsoids? I know I originally asked about axis and shaded octants but after using the anisotropic option for the last month or so I notice tat there is a tremendous amount of information conveyed with just the color and shape and shaded octants may not add much information (in fact it may just confuse the issue). Your implementation of thermal ellipsoids is very, very useful! 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 Thu, 6 Aug 2009, Michael Day wrote:
How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid.
Cheers, Mike
In the next daily build, we will have interactive label moving, where you can pick a label and drag it around, and if you hold down the shift key, it will move in Z. It will default to the Control-button-3 mouse mode, but can be reassigned using the mouse mode preferences. You will also be able to change the label offsets using the command line, the "labeloffset x y z atom-spec" and "rlabeloffset x y z atom-spec" commands to set the offsets and "~labeloffset" and "~rlabeloffset" to reset them to the default behavior of adjusting the offset with the representation. For users with existing saved preferences, you might need to turn label dragging on by resetting the mouse mode preferences or by explicitly assigning it to Control-button-3. There's no icon for it yet in the mouse modes interface, but it will be the rightmost column for now. Since the label moving code is new, please send me feedback about how well it works for you. - Greg
Greg and Eric, Thank you. I installed the build and it works great. The default offset when using thermal ellipsoids is very nice. I haven't figured out how to drag individual labels but that may be me or that I'm using a trackball, but the labeloffset command works well. I'm getting very positive feedback from my users! Thanks. New question, Is it possible to calculate electron density maps from a standard PDB format fcf file with the structure factors in it? SHELX is the small molecule standard and it will output reflection files. I know COOT will calculate the maps on the fly from this format but figures from COOT are NO MATCH for Chimera. It would be nice to be able to read the .fcf file and get maps back. If I try to use COOT calculated maps I have no control over the position of the map and it can be the same for trying to use CCP4 maps. Maps calculated in Chimera centered on the molecule would be useful beyond description. From the SHELX manual m = 6:Write a free-format CIF file containing h,k,l, Fo, σ(Fo), Fc and φ (phase angle in degrees) for the reflection list as defined for m = 1. This is the recommended format for the deposition of reflection data with the PDB, and is also the format required for the generation of refinement statistics and electron density maps using SHELXPRO An actal file # # h,k,l, Fo-squared, sigma(Fo-squared), Fc and phi(calc) # data_ta36 _shelx_title ' ta36 in P-1' _shelx_refln_list_code 6 _shelx_F_calc_maximum 147.61 _exptl_crystal_F_000 628.00 _reflns_d_resolution_high 0.7205 loop_ _symmetry_equiv_pos_as_xyz 'x, y, z' '-x, -y, -z' _cell_length_a 10.9827 _cell_length_b 11.4038 _cell_length_c 12.1961 _cell_angle_alpha 65.918 _cell_angle_beta 73.507 _cell_angle_gamma 76.655 loop_ _refln_index_h _refln_index_k _refln_index_l _refln_F_squared_meas _refln_F_squared_sigma _refln_F_calc _refln_phase_calc 1 0 0 548.13 8.13 23.91 0.0 2 0 0 713.05 6.90 27.04 0.0 3 0 0 4487.70 36.52 61.90 0.0 4 0 0 5185.71 41.46 70.98 0.0 And the relevant part of the corresponding SHELX ins file TITL ta36 in P-1 CELL 0.71073 10.9827 11.4038 12.1961 65.918 73.507 76.655 ZERR 1.00 0.0005 0.0006 0.0006 0.003 0.003 0.003 LATT 1 SFAC C H N O F S Cu UNIT 52 45 9 8 6 2 2 WGHT 0.000000 FVAR 0.66533 0.53147 CU1 7 -0.011240 0.599643 0.665074 11.00000 0.02325 0.03784 = 0.02594 -0.00922 -0.00439 -0.00980 O1 4 0.323760 0.794114 0.628981 11.00000 0.02076 0.03749 = 0.07654 -0.03652 -0.01288 -0.00180 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 Aug 7, 2009, at 2:44 PM, Greg Couch wrote: On Thu, 6 Aug 2009, Michael Day wrote:
How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid.
Cheers, Mike
In the next daily build, we will have interactive label moving, where you can pick a label and drag it around, and if you hold down the shift key, it will move in Z. It will default to the Control-button-3 mouse mode, but can be reassigned using the mouse mode preferences. You will also be able to change the label offsets using the command line, the "labeloffset x y z atom-spec" and "rlabeloffset x y z atom- spec" commands to set the offsets and "~labeloffset" and "~rlabeloffset" to reset them to the default behavior of adjusting the offset with the representation. For users with existing saved preferences, you might need to turn label dragging on by resetting the mouse mode preferences or by explicitly assigning it to Control-button-3. There's no icon for it yet in the mouse modes interface, but it will be the rightmost column for now. Since the label moving code is new, please send me feedback about how well it works for you. - Greg
Hi Michael, How difficult would it be for Chimera to compute x-ray maps from structure factors? For deposited Protein Databank models with structure factors, the Uppsala Electron Density Server (EDS) provides maps. http://eds.bmc.uu.se/eds/ These maps can be fetched by Chimera (File / Fetch by ID). The procedure the describe for computing 2fo-fc maps is quite intricate http://eds.bmc.uu.se/eds/eds_help.html#NITTY_GRITTY They are trying to verify the reported R-factor and do other sanity checks. Still it may be too complex a procedure even without the quality controls for us to put into Chimera. But we don't know much about crystallographic refinement software. Maybe you could provide more details to show us that it is in fact easy to compute the maps? It may be there are easy solutions to get externally generated maps to align with the models in Chimera. Both COOT and CCP4 know how to correctly align the PDB models and maps they calculate so that should be possible in Chimera too. I am not very knowledgable about COOT or CCP4 and any information about the alignment they use that you could provide might help us improve use of those maps in Chimera. Tom -------- Original Message -------- Subject: Re: [Chimera-users] Moving atom labels From: Michael Day <mikeday@caltech.edu> To: Greg Couch <gregc@cgl.ucsf.edu>, Eric Pettersen <pett@cgl.ucsf.edu> Date: 8/10/09 6:05 PM
Greg and Eric,
Thank you. I installed the build and it works great. The default offset when using thermal ellipsoids is very nice. I haven't figured out how to drag individual labels but that may be me or that I'm using a trackball, but the labeloffset command works well.
I'm getting very positive feedback from my users!
Thanks.
New question,
Is it possible to calculate electron density maps from a standard PDB format fcf file with the structure factors in it? SHELX is the small molecule standard and it will output reflection files. I know COOT will calculate the maps on the fly from this format but figures from COOT are NO MATCH for Chimera. It would be nice to be able to read the .fcf file and get maps back. If I try to use COOT calculated maps I have no control over the position of the map and it can be the same for trying to use CCP4 maps. Maps calculated in Chimera centered on the molecule would be useful beyond description. * * *From the SHELX manual* *m = 6*:Write a free-format CIF file containing h,k,l, Fo, σ(Fo), Fc and φ (phase angle in degrees) for the reflection list as defined for m = 1. This is the recommended format for the deposition of reflection data with the PDB, and is also the format required for the generation of refinement statistics and electron density maps using SHELXPRO
*An actal file* # # h,k,l, Fo-squared, sigma(Fo-squared), Fc and phi(calc) # data_ta36 _shelx_title ' ta36 in P-1' _shelx_refln_list_code 6 _shelx_F_calc_maximum 147.61 _exptl_crystal_F_000 628.00 _reflns_d_resolution_high 0.7205
loop_ _symmetry_equiv_pos_as_xyz 'x, y, z' '-x, -y, -z'
_cell_length_a 10.9827 _cell_length_b 11.4038 _cell_length_c 12.1961 _cell_angle_alpha 65.918 _cell_angle_beta 73.507 _cell_angle_gamma 76.655
loop_ _refln_index_h _refln_index_k _refln_index_l _refln_F_squared_meas _refln_F_squared_sigma _refln_F_calc _refln_phase_calc 1 0 0 548.13 8.13 23.91 0.0 2 0 0 713.05 6.90 27.04 0.0 3 0 0 4487.70 36.52 61.90 0.0 4 0 0 5185.71 41.46 70.98 0.0�
*And the relevant part of the corresponding SHELX ins file* TITL ta36 in P-1 CELL 0.71073 10.9827 11.4038 12.1961 65.918 73.507 76.655 ZERR 1.00 0.0005 0.0006 0.0006 0.003 0.003 0.003 LATT 1 SFAC C H N O F S Cu UNIT 52 45 9 8 6 2 2
WGHT 0.000000 FVAR 0.66533 0.53147 CU1 7 -0.011240 0.599643 0.665074 11.00000 0.02325 0.03784 = 0.02594 -0.00922 -0.00439 -0.00980 O1 4 0.323760 0.794114 0.628981 11.00000 0.02076 0.03749 = 0.07654 -0.03652 -0.01288 -0.00180�
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 <mailto:mikeday@caltech.edu> <<< ------------------------------------------------------------------------>>>
On Aug 7, 2009, at 2:44 PM, Greg Couch wrote:
On Thu, 6 Aug 2009, Michael Day wrote:
How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid.
Cheers, Mike
In the next daily build, we will have interactive label moving, where you can pick a label and drag it around, and if you hold down the shift key, it will move in Z. It will default to the Control-button-3 mouse mode, but can be reassigned using the mouse mode preferences. You will also be able to change the label offsets using the command line, the "labeloffset x y z atom-spec" and "rlabeloffset x y z atom-spec" commands to set the offsets and "~labeloffset" and "~rlabeloffset" to reset them to the default behavior of adjusting the offset with the representation.
For users with existing saved preferences, you might need to turn label dragging on by resetting the mouse mode preferences or by explicitly assigning it to Control-button-3. There's no icon for it yet in the mouse modes interface, but it will be the rightmost column for now.
Since the label moving code is new, please send me feedback about how well it works for you.
- Greg
------------------------------------------------------------------------
_______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
On Fri, 7 Aug 2009, Greg Couch wrote:
On Thu, 6 Aug 2009, Michael Day wrote:
How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid.
Cheers, Mike
In the next daily build, we will have interactive label moving, where you can pick a label and drag it around, and if you hold down the shift key, it will move in Z. It will default to the Control-button-3 mouse mode, but can be reassigned using the mouse mode preferences. You will also be able to change the label offsets using the command line, the "labeloffset x y z atom-spec" and "rlabeloffset x y z atom-spec" commands to set the offsets and "~labeloffset" and "~rlabeloffset" to reset them to the default behavior of adjusting the offset with the representation.
For users with existing saved preferences, you might need to turn label dragging on by resetting the mouse mode preferences or by explicitly assigning it to Control-button-3. There's no icon for it yet in the mouse modes interface, but it will be the rightmost column for now.
Since the label moving code is new, please send me feedback about how well it works for you.
- Greg
In the next daily build, the labeloffset and rlabeloffset commands have been replaced with an offset argument to the label and rlabel commands: 'label' ['offset' ('default' | x,y,z)] atom-spec and likewise for rlabel. [] designates optional. (|) designates choices. '' designates literal text. x,y,z is three numbers separated by commas and no embedded spaces. The default offset resets to the default behavior, like ~labeloffset before. - Greg
Hi Mike, On Aug 6, 2009, at 6:02 PM, Michael Day wrote:
How does one change the position of atom labels? For small coordination complexes the default atom labels land on the atom and it would be nice to move them as a whole or individually so they are alongside the atom ball or anisotropic ellipsoid.
For the 1.4 release we hope to have label repositioning with the mouse available (as listed on this Wiki page: http://socrates2.cgl.ucsf.edu/trac/chimera/wiki/Release1.4) . We don't have it yet. However, the underlying mechanism is in place and I have modified Thermal Ellipsoids to offset atom labels by an amount equal to the longest half-axis of the atom's displayed ellipsoid. That will be in tomorrow's build. Furthermore, I made it so that you can set the label-offset attribute with the "setattr" command. This required some mods to that command since the label-offset attribute is a vector and the setattr command didn't support non-scalar values. So in tomorrow's build this command will move selected atom's labels half an angstrom to the right: setattr a labelOffset 0.5,0,0 sel It's not 100% certain that this is how setting vector attributes will work in the 1.4 release, but if it changes I'll let people know.
While I'm on the ellipsoid topic; has there been progress on overlaying axis on the ellipsoids? I know I originally asked about axis and shaded octants but after using the anisotropic option for the last month or so I notice tat there is a tremendous amount of information conveyed with just the color and shape and shaded octants may not add much information (in fact it may just confuse the issue).
Well....no, not really. I've had a lot of my time tied up with system administration issues related to a new Linux cluster we're converting over to (principally getting an efficient reliable backup system implemented), but the good news is that that time commitment is winding down and I will have more time for issues such as these that I would prefer to work on. So hopefully soon. The only thing I've managed to do is get some of the underlying infrastructure for the probability calculator done (basically conscripting a Simpson's Rule numerical integrator once I understood what I needed).
Your implementation of thermal ellipsoids is very, very useful!
Thanks! --Eric Eric Pettersen UCSF Computer Graphics Lab
On Aug 7, 2009, at 2:51 PM, Eric Pettersen wrote:
Furthermore, I made it so that you can set the label-offset attribute with the "setattr" command.
Ignore this. Use the commands Greg implemented for now (assuming for some reason that you wanted to use the command version). Those commands will likely change before the production release. Specifically, they will likely be incorporated as options to the label/ rlabel commands. --Eric Eric Pettersen USCF Computer Graphics Lab
participants (4)
-
Eric Pettersen -
Greg Couch -
Michael Day -
Tom Goddard