Hi Oli,

Ok I added a "move cofr <models-spec>" command, for example,

  move cofr #2

Syntax is slightly different than in Chimera which required a "models" keyword.  Also the distance and frames options seen in the Chimera command are not in ChimeraX.  If multiple models are specified they are moved rigidly as a group so that their bounding box center (displayed parts only) is moved to the current center of rotation.

Tom

From: Oliver Clarke <olibclarke@gmail.com>
Subject: Re: [ChimeraX] #14408: Add command to move models to center of rotation
Date: January 18, 2024 at 4:31:57 PM PST
To: ChimeraX-bugs@cgl.ucsf.edu
Cc: goddard@cgl.ucsf.edu

Yes an option for fitmap could do the trick, although a separate move cofr command would be more flexible in my opinion.

In Chimera, I would generally move the cofr where I want to position the model, “move cofr” to move it, then I would either use fitmap with global search (“search 100 radius 5”) or quickly manually reorient (which I do about equally as often). It might sound cumbersome but feels pretty quick and intuitive!

Either way, I find being able to reposition a model to the center of rotation useful.

Cheers
Oli


On Jan 18, 2024, at 4:12 PM, Tom Goddard <goddard@sonic.net> wrote:

Hi Oli,

 I don’t quite understand how this is useful.  If you position a model at the center of rotation don’t you also have to rotate it to a sensible orientation.  How do you do the rotation?  Are you trying to do a fit in a map?  I agree the current way that is done in ChimeraX where you have to move the map by hand is cumbersome.  Maybe instead an option to the fitmap command to first move the model to the center of rotation and then optimize the fit (possibly with a rotational search) would better get at what are trying to do.

Tom

On Jan 18, 2024, at 3:14 PM, Oliver Clarke via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Would it be possible to add a "move cofr" command to chimerax at some point? We use this frequently in Chimera to move an atomic model to a precise point in a map (when we often don't have any other models in there to align to).

I know one can do this graphically using "translate selected models", but it is much easier and more precise in my experience to position the center of rotation (using the showPivot indicator) where I want to be, and then move the model to the center of rotation.

Cheers
Oli

On Jan 17, 2024, at 3:48 PM, Tom Goddard <goddard@sonic.net> wrote:

Hi Oli,

There is no command in ChimeraX that moves models to the center of rotation like the Chimera "move cofr" command.  If the idea is to use the shape command and then place the shape at the center of an existing map (e.g. model #1) then I would suggest using the "center" option of the shape command

shape cylinder radius 50 height 100 center #1

It is not obvious where the center of rotation point is in ChimeraX, for instance it may be on the surface of a displayed map in the center of the view that is not near the center of the map.  So I don't think a "move cofr" would do this job very well.

Tom


On Jan 17, 2024, at 10:54 AM, Oliver Clarke <olibclarke@gmail.com> wrote:

Relatedly - is there any ChimeraX command line equivalent to "move cofr" in Chimera - to move a designated model to the current center of rotation?

Cheers
Oli

On Jan 17, 2024, at 1:43 PM, Oliver Clarke <olibclarke@gmail.com> wrote:

Perfect thanks Tom - I did not know about the "volume onesmask" command (I looked in the docs but clearly not in the right place!), but that is exactly what I was looking for!

Cheers
Oli

On Jan 17, 2024, at 1:39 PM, Tom Goddard <goddard@sonic.net> wrote:

Hi Oliver,

This is easy to do with the existing "volume onesmask" and "volume falloff" commands.  For example

open 1080 from emdb
shape cylinder radius 50 height 100
volume onesmask #2 ongrid #1
volume falloff #3

I don't think it is worthwhile to cram all the onesmask and falloff options into the shape command.  You might also use the "volume gaussian" command instead of the "volume falloff" command to soften the edges.  The volume falloff command will not modifiy the ones region, only the zeros outside the boundary, while the volume gaussian command will modify inside and outside the mask.

Tom


On Jan 17, 2024, at 8:22 AM, Oliver Clarke via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Hi,

Would it be possible in a future release to add an option to the "shape" command to output the shape as a map (filled with 1s inside the shape, zeroes outside, maybe with an optional soft edge and an ongrid option)?

This would be very convenient for making simple geometrical masks for cryoEM (e.g soft cylindrical masks for classification within a nanodisc).

I currently do this using "shape cylinder", use the cylinder to mask the map, and then gaussian filter the resulting masked map to the point where I can choose a threshold that conforms to the shape, but having a way to do the same thing more directly would be very nice to have.

Cheers
Oli
_______________________________________________
ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu
To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu
Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/








_______________________________________________
ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu
To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu
Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/