I'm putting together a demo for my students this next semester, and was hoping to be able to do a zone selection about the ligand. Is this possible with the command line? I haven't been able to pull out a reference for such, yet.
Certainly I could pull together the list of residues myself, but thought it'd be nice to have the command available. Some students might like to create a demo file for their reports, and this would become a pivotal tool for a subset of those.
If not directly available, then a wishlist addition, perhaps? Something like:
zonesel [within|beyond] [+/- angstroms] [atom-spec]
where the sign of the angstrom value denotes whether to include (+) or exclude (-) the atoms in atom-spec.
Cheers,
Kenward Vaughan
Hi Kenward, There *is* a command-line way to specify atoms/residues in a zone defined by other atoms. It might have been hard to find since it's not a separate command, but instead an atom specification that can be used with many commands.
The general form is command atom-spec za<N for example select ligand za<3.5
to select all atoms within 3.5 angstroms of any ligand atoms. Analogously, "za>N" could be used for all atoms farther away than N angstroms. Using "z" or "zr" instead of "za" gives a residue-based cutoff - you would always get whole residues even if only parts of them fall in the zone. The "command" can be "select" or many other things (color, display, etc.).
Here are a couple more examples: rlabel ions z<4 (label residues within 4 angstroms of any ions) disp #0 & #2 z<5 (display residues in model 0 that are within 5 angstroms of any atoms in model 2).
The description in the manual (click the "zones" link): http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/ frameatom_spec.html
It can be confusing, so feel free to write back if you have problems generating the command needed in your particular situation! Best, Elaine ----- Elaine C. Meng, Ph.D. meng@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 Dec 30, 2006, at 8:16 AM, Kenward Vaughan wrote:
I'm putting together a demo for my students this next semester, and was hoping to be able to do a zone selection about the ligand. Is this possible with the command line? I haven't been able to pull out a reference for such, yet.
Certainly I could pull together the list of residues myself, but thought it'd be nice to have the command available. Some students might like to create a demo file for their reports, and this would become a pivotal tool for a subset of those.
If not directly available, then a wishlist addition, perhaps? Something like:
zonesel [within|beyond] [+/- angstroms] [atom-spec]
where the sign of the angstrom value denotes whether to include (+) or exclude (-) the atoms in atom-spec.
Cheers,
Kenward Vaughan
.'^~;,_Dr. Kenward Vaughan `:,'~~~~~ Professor of Chemistry ;:/ Bakersfield College |,;| 1801 Panorama Drive / ', \ Bakersfield, CA 93305 / o O \ http://www2.bc.cc.ca.us/kvaughan (oOoOOoOo) ---========--- ???$$MM$$???
Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
On Sat, 30 Dec 2006, Kenward Vaughan wrote:
I'm putting together a demo for my students this next semester, and was hoping to be able to do a zone selection about the ligand. Is this possible with the command line? I haven't been able to pull out a reference for such, yet.
Certainly I could pull together the list of residues myself, but thought it'd be nice to have the command available. Some students might like to create a demo file for their reports, and this would become a pivotal tool for a subset of those.
If not directly available, then a wishlist addition, perhaps? Something like:
zonesel [within|beyond] [+/- angstroms] [atom-spec]
where the sign of the angstrom value denotes whether to include (+) or exclude (-) the atoms in atom-spec.
Take a look at the Atom Specification documentation in the subsection on Zones. So:
display ligand z<5
Will display all of the residues that are less than 5 angstroms from the ligand.
- Greg
On Saturday 30 December 2006 10:09, you wrote:
On Sat, 30 Dec 2006, Kenward Vaughan wrote:
I'm putting together a demo for my students this next semester, and was hoping to be able to do a zone selection about the ligand. Is this possible with the command line? I haven't been able to pull out a reference for such, yet.
Certainly I could pull together the list of residues myself, but thought it'd be nice to have the command available. Some students might like to create a demo file for their reports, and this would become a pivotal tool for a subset of those.
If not directly available, then a wishlist addition, perhaps? Something like:
zonesel [within|beyond] [+/- angstroms] [atom-spec]
where the sign of the angstrom value denotes whether to include (+) or exclude (-) the atoms in atom-spec.
Take a look at the Atom Specification documentation in the subsection on Zones. So:
display ligand z<5
Will display all of the residues that are less than 5 angstroms from the ligand.
- Greg
Ah. That was too easy! I obviously didn't go deep enough into that section. :( Many thanks, Greg!
Kenward
On Saturday 30 December 2006 10:08, you wrote:
Hi Kenward, There *is* a command-line way to specify atoms/residues in a zone defined by other atoms. It might have been hard to find since it's not a separate command, but instead an atom specification that can be used with many commands.
The general form is command atom-spec za<N
...
It can be confusing, so feel free to write back if you have problems generating the command needed in your particular situation! Best, Elaine
Thank you so much, Elaine and Greg. I appreciate the fact that y'all are up and running even today (things are different at a community college :).
In my demo, I've found that rotating a model about to create a specific view in one panel, followed in the next panel by either changing the atom representation or recoloring the atoms, causes a resetting of the view to the opening orientation. Is this a known issue?
Kenward
ps. BTW, can a selected group of atoms be named at the command line? Another question I can't seem to find an answer to... :(
Hi Kenward, Just changing the color or representation should not reset the orientation *achieved by commands* (more explanation below) at the end of the previous panel. There are examples in the demos that come with Chimera (Tools... Demos) - e.g. in the hormone-receptor demo, the position does not reset to the original when switching from panel 2 to panel 3 (FSH undisplayed, leucines in FSH receptor displayed). I also tried playing this demo up to panel 2, opening the demo in the editor (from the Demo dialog, File... Open in Editor), unchecking the roll commands in panel 3, and then clicking Execute to test whether having some movement suppresses a reset bug. However, there was still no reset between panels 2 and 3. I'm using version 1.2304.
Note that any rotations/translations performed *manually* will not be incorporated into a panel, only the ones expressed as commands. If you rotated/translated manually, there will be a resetting to the orientation without those manual movements before the next set of commands are applied. Perhaps that is what happened. That is the tricky thing about Demos: everything must be done with commands, and it can take a fair amount of effort to get the orientations you want. Actually, the resetting to the orientation achieved purely by commands is at least partly a feature rather than a bug, because it allows demo viewers to change the view manually within a panel however they desire without disrupting the views the demo creator intended to be visited during the course of the demo. Hope that makes sense... It may be instructive to open the existing demos in the editor as mentioned above, if you haven't tried that already.
If the problem persists, let me know what version of Chimera you're using. Also, if you want you could send me your demo file and any associated data files and I could take a more in-depth look next week.
About specifying a bunch of atoms over and over, you can name a selection and then use the selection's name. See http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/menu.html#namesel
However, a selection can only be named using the menu currently, so for Demos a different approach is required. In similar situations, I bypass selection entirely and instead alias that set of atoms (using the same specifications that you would have used to select the atoms) and then use the resulting alias in subsequent commands. For example:
Command: alias key #0:17.a,138.a,190.a #1:38.b,158.b,221.b #2:86.a,247.a,342.a Command: rep sphere key Command: focus key Command: color byatom key
The alias statement can include all kinds of specifications - zones, categories such as "ligand," combination operators & | ~ , etc. Man page for "alias": http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/alias.html
Technically we aren't working today. However, I like to check my mail every day and will occasionally answer mail from devoted Chimera users! 8-)
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. meng@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 Dec 30, 2006, at 12:30 PM, Kenward Vaughan wrote:
In my demo, I've found that rotating a model about to create a specific view in one panel, followed in the next panel by either changing the atom representation or recoloring the atoms, causes a resetting of the view to the opening orientation. Is this a known issue?
Kenward
ps. BTW, can a selected group of atoms be named at the command line? Another question I can't seem to find an answer to... :( -- .'^~;,_ Dr. Kenward Vaughan `:,'~~~~~ Professor of Chemistry ;:/ Bakersfield College |,;| 1801 Panorama Drive / ', \ Bakersfield, CA 93305 / o O \ http://www2.bc.cc.ca.us/kvaughan (oOoOOoOo) ---========--- ???$$MM$$???
On Saturday 30 December 2006 14:29, Elaine Meng wrote:
Hi Kenward, Just changing the color or representation should not reset the orientation *achieved by commands* (more explanation below) at the end of the previous panel. There are examples in the demos that come with Chimera (Tools... Demos) - e.g. in the hormone-receptor demo, the position does not reset to the original when switching from panel 2 to panel 3 (FSH undisplayed, leucines in FSH receptor displayed). I also tried playing this demo up to panel 2, opening the demo in the editor (from the Demo dialog, File... Open in Editor), unchecking the roll commands in panel 3, and then clicking Execute to test whether having some movement suppresses a reset bug. However, there was still no reset between panels 2 and 3. I'm using version 1.2304.
...
If the problem persists, let me know what version of Chimera you're using. Also, if you want you could send me your demo file and any associated data files and I could take a more in-depth look next week.
I have uncovered the "problem." In one panel, I might have the following:
roll y 1 22 ; roll x -0.8182 22
and in the following panel:
repre stick ; color byatom
This causes the view in the second panel to revert to what existed prior to the first. **This happens regardless of how long one waits before moving forward from one panel to the next. It is independent of any manual manipulation as well.**
If the first line is terminated with a wait:
roll y 1 22 ; roll x -0.8182 22 ; wait
then the view generated by the rolls is maintained.
I took the COX inhibitor demo and tested this with several lines in there as well (7-10, IIRC). The results matched what I see with my own. I've attached two small jpg's which show the effect of removing the wait found in panel 9.
I had understood from the help files that "wait" insured following commands were not processed until molecular motion had stopped. I had not included it in my lines since I knew that it wouldn't be needed (the rolls were the only commands processed before control was relinquished back to the user).
Apparently the "wait" is required to properly set the coordinates for the next panel.
About specifying a bunch of atoms over and over, you can name a selection and then use the selection's name. See http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/menu.html#namesel
However, a selection can only be named using the menu currently, so for Demos a different approach is required. In similar situations, I bypass selection entirely and instead alias that set of atoms (using the same specifications that you would have used to select the atoms)
...
It's definitely a good start! Thanks!
Technically we aren't working today. However, I like to check my mail every day and will occasionally answer mail from devoted Chimera users! 8-)
I hope this helps, Elaine
Happy New Year! :-)
Kenward
On Dec 30, 2006, at 12:30 PM, Kenward Vaughan wrote:
In my demo, I've found that rotating a model about to create a specific view in one panel, followed in the next panel by either changing the atom representation or recoloring the atoms, causes a resetting of the view to the opening orientation. Is this a known issue?
Kenward
ps. BTW, can a selected group of atoms be named at the command line? Another question I can't seem to find an answer to... :(
On Saturday 30 December 2006 14:29, Elaine Meng wrote:
Hi Kenward, Just changing the color or representation should not reset the orientation *achieved by commands* (more explanation below) at the end of the previous panel. There are examples in the demos that come with Chimera (Tools... Demos) - e.g. in the hormone-receptor demo, the position does not reset to the original when switching from panel 2 to panel 3 (FSH undisplayed, leucines in FSH receptor displayed). I also tried playing this demo up to panel 2, opening the demo in the editor (from the Demo dialog, File... Open in Editor), unchecking the roll commands in panel 3, and then clicking Execute to test whether having some movement suppresses a reset bug. However, there was still no reset between panels 2 and 3. I'm using version 1.2304.
...
If the problem persists, let me know what version of Chimera you're using. Also, if you want you could send me your demo file and any associated data files and I could take a more in-depth look next week.
I have uncovered the "problem." In one panel, I might have the following:
roll y 1 22 ; roll x -0.8182 22
and in the following panel:
repre stick ; color byatom
This causes the view in the second panel to revert to what existed prior to the first. **This happens regardless of how long one waits before moving forward from one panel to the next. It is independent of any manual manipulation as well.**
If the first line is terminated with a wait:
roll y 1 22 ; roll x -0.8182 22 ; wait
then the view generated by the rolls is maintained.
I took the COX inhibitor demo and tested this with several lines in there as well (7-10, IIRC). The results matched what I see with my own. I've attached two small jpg's which show the effect of removing the wait found in panel 9.
I had understood from the help files that "wait" insured following commands were not processed until molecular motion had stopped. I had not included it in my lines since I knew that it wouldn't be needed (the rolls were the only commands processed before control was relinquished back to the user).
Apparently the "wait" is required to properly set the coordinates for the next panel.
About specifying a bunch of atoms over and over, you can name a selection and then use the selection's name. See http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/menu.html#namesel
However, a selection can only be named using the menu currently, so for Demos a different approach is required. In similar situations, I bypass selection entirely and instead alias that set of atoms (using the same specifications that you would have used to select the atoms)
...
It's definitely a good start! Thanks!
Technically we aren't working today. However, I like to check my mail every day and will occasionally answer mail from devoted Chimera users! 8-)
I hope this helps, Elaine
Happy New Year! :-)
Kenward
On Dec 30, 2006, at 12:30 PM, Kenward Vaughan wrote:
In my demo, I've found that rotating a model about to create a specific view in one panel, followed in the next panel by either changing the atom representation or recoloring the atoms, causes a resetting of the view to the opening orientation. Is this a known issue?
Kenward
ps. BTW, can a selected group of atoms be named at the command line? Another question I can't seem to find an answer to... :(
On Dec 30, 2006, at 7:14 PM, Kenward Vaughan wrote:
I have uncovered the "problem." In one panel, I might have the following:
roll y 1 22 ; roll x -0.8182 22
and in the following panel:
repre stick ; color byatom
This causes the view in the second panel to revert to what existed prior to the first. **This happens regardless of how long one waits before moving forward from one panel to the next. It is independent of any manual manipulation as well.**
If the first line is terminated with a wait:
roll y 1 22 ; roll x -0.8182 22 ; wait
then the view generated by the rolls is maintained.
Hi Kenward, Thanks for noticing this! It is somewhat rationalizable, because the "savepos" command (save position) is executed after each set of panel commands. If there isn't a "wait" at the end, the position is will be saved prematurely and not reflect all panel commands. However, you are right, it should be smarter and automatically incorporate a "wait" instead of putting the onus on the demo creator.
We will file this as a bug. In the meanwhile, using "wait" after the panel commands is the way to go. I never noticed this issue, because to make a demo I first created a single script (chock full of "wait" commands) and then broke it up into panels.
I would be interested to see how your demo turns out! Please don't hesitate to contact us if any other issues arise. Best, Elaine
On Dec 30, 2006, at 12:30 PM, Kenward Vaughan wrote:
ps. BTW, can a selected group of atoms be named at the command line? Another question I can't seem to find an answer to... :(
Elaine gives good suggestions for workarounds, nonetheless this should be possible. I'll add it to the list of requested features.
--Eric
Eric Pettersen UCSF Computer Graphics Lab pett@cgl.ucsf.edu http://www.cgl.ucsf.edu
participants (4)
-
Elaine Meng -
Eric Pettersen -
Greg Couch -
Kenward Vaughan