[chimerax-users] A question about model position matrix

[Zheng Ruan]

Thank you, Elaine and Tom. This makes sense. I will center my model first
such that the position matrix will be more interpretable.

Best,
Zheng

On Thu, Mar 31, 2022 at 2:04 PM Tom Goddard <goddard at sonic.net> wrote:

> Hi Zheng,
>
>   Here's a bit more explanation.  Each model has a coordinate system and
> the ChimeraX scene has its own coordinate system.  When you first load a
> model its coordinate system is the same as the scene coordinate system.
> But if you move the model using for instance the "model" option of the turn
> command it will move the model within the scene and the model and scene
> coordinate systems will be different.
>
>   The turn command by default uses a center of rotation which is the
> center of the bounding box of the displayed part of your model.  That
> center will not be 0,0,0 in general.  So when you used the turn command
> without specifying the center, it rotated about a point that was not
> 0,0,0.  When you use the "view matrix" command it reports the position of
> models as a rotation around 0,0,0 followed by a translation.  Your rotation
> done with the turn command was not centered on 0,0,0 so it is equivalent to
> a rotation about 0,0,0 plus some non-zero translation, which is what you
> observed.
>
>   So as Elaine points out you need to specify "center 0,0,0" in your turn
> command.
>
>   One further detail.  The rotation and translation reported by view
> matrix maps a point in model coordinates to scene coordinates.  You might
> think it maps from scene to model coordinates but that would require
> inverting the transformation.
>
>         Tom
>
>
> > On Mar 31, 2022, at 9:40 AM, Elaine Meng via ChimeraX-users <
> chimerax-users at cgl.ucsf.edu> wrote:
> >
> > Hi Zheng,
> > My mistake, instead of "center #1" I should have said "center 0,0,0" ...
> try that:
> >
> > turn z 90 models #1 center 0,0,0
> >
> > Then the translation should be all zero.
> > Best,
> > Elaine
> > -----
> > Elaine C. Meng, Ph.D.
> > UCSF Chimera(X) team
> > Department of Pharmaceutical Chemistry
> > University of California, San Francisco
> >
> >> On Mar 31, 2022, at 9:30 AM, Zheng Ruan via ChimeraX-users <
> chimerax-users at cgl.ucsf.edu> wrote:
> >>
> >> Hi Elaine,
> >>
> >> Thank you for the reply. However, I don't see a large translation of my
> map after I type the "turn" command. I only have one map loaded in ChimeraX
> and the map is rotated about its own z-axis.
> >>
> >> I also tried to specify the "center" option as suggested, but I still
> see a large translation value. For example,
> >>
> >> open volume.mrc format mrc
> >>
> >> view matrix
> >> camera position: 1,0,0,345.63,0,1,0,343.7,0,0,1,1042.6
> >> model positions: #1,1,0,0,0,0,1,0,0,0,0,1,0,#1.1,1,0,0,0,0,1,0,0,0,0,1,0
> >>
> >> turn z 90 models #1 center #1
> >>
> >> view matrix
> >> camera position: 1,0,0,345.63,0,1,0,343.7,0,0,1,1042.6
> >> model positions:
> #1,1.1102e-16,-1,0,689.33,1,1.1102e-16,0,-1.9358,0,0,1,0,#1.1,1,0,0,0,0,1,0,0,0,0,1,0
> >>
> >> I also tried to change the center of rotation, but the result remains
> the same.
> >>
> >> Best,
> >> Zheng
> >>
> >>
> >> On Thu, Mar 31, 2022 at 11:20 AM Elaine Meng <meng at cgl.ucsf.edu> wrote:
> >> Dear Zheng,
> >> I believe it is because the center of rotation from your command is not
> the same as the center of the model.  If the center of rotation is far from
> the model, it might be translated a large distance in addition to being
> rotated.  You can specify the center of rotation for the "turn" command
> with the "center" option.  I believe that if you use
> >>
> >> turn z 90 models #1 center #1
> >>
> >> ...the translation of the model would then be zero or close to zero.
> >>
> >> See help for "turn"
> >> <https://rbvi.ucsf.edu/chimerax/docs/user/commands/turn.html>
> >>
> >> I hope this helps,
> >> Elaine
> >> -----
> >> Elaine C. Meng, Ph.D.
> >> UCSF Chimera(X) team
> >> Department of Pharmaceutical Chemistry
> >> University of California, San Francisco
> >>
> >>
> >>> On Mar 30, 2022, at 8:18 PM, Zheng Ruan via ChimeraX-users <
> chimerax-users at cgl.ucsf.edu> wrote:
> >>>
> >>> Dear ChimeraX users,
> >>>
> >>> I'm trying to understand the model position matrix. According to the
> documentation, "Each matrix is reported as 12 comma-separated numbers,
> corresponding to a 3x3 rotation matrix and a translation vector in the
> fourth column. Ordering is row-by-row, such that the translation vector is
> given as the fourth, eighth, and twelfth numbers."
> >>>
> >>> When I load a model into ChimeraX and use "view matrix" command to
> check the model position, I got
> "#1,1,0,0,0,0,1,0,0,0,0,1,0,#1.1,1,0,0,0,0,1,0,0,0,0,1,0". The rotation
> matrix for #1 is an identity matrix with translation vector to be 0 for all
> directions.
> >>>
> >>> However, if I rotate the model using the command "turn z 90 models #1"
> and then check the model position using "view matrix", I got something like
> "#1,1.1102e-16,-1,0,689.33,1,1.1102e-16,0,-1.9358,0,0,1,0,#1.1,1,0,0,0,0,1,0,0,0,0,1,0".
> The translation vector for #1 becomes (689.33, -1.9358, 0).
> >>>
> >>> Since I only applied a simple rotation of the model along z-axis, why
> the translation vector becomes non-zero in this case? I hope to have a
> better understanding of how ChimeraX handles the model position matrix to
> describe model positions.
> >>>
> >>> Thank you so much!
> >>> Best,
> >>> Zheng
> >>
> >> _______________________________________________
> >> ChimeraX-users mailing list
> >> ChimeraX-users at cgl.ucsf.edu
> >> Manage subscription:
> >> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
> >
> >
> > _______________________________________________
> > ChimeraX-users mailing list
> > ChimeraX-users at cgl.ucsf.edu
> > Manage subscription:
> > https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rbvi.ucsf.edu/pipermail/chimerax-users/attachments/20220331/b48fd1d7/attachment.html>