Dual view windows to compare structures

[mheller@…]

Begin forwarded message:

From: Markus Heller
Subject: [Chimera-users] Split view feature
Date: September 30, 2016 at 10:41:32 PM PDT
To: chimera List <chimera-users@…>

Hi all,

Not sure split view is the proper term for what I’m looking for but here goes: I want to compare to structures of two proteins bound to their respective ligands, and I do NOT want to do this as overlay. What I want is a split screen or split view in the Chimera main window, where one structure is displayed on the left, and the other on the right. Movements (rotation, translation etc.) would be synchronized between both windows, whereas other functionality such as Zone select, find Hbonds etc. would not.

Is such a feature available for the current release, or is it planned for ChimeraX?

Thanks and Cheers
Markus

[Tom Goddard]

I like your dual view idea for ChimeraX. It is a common situation that you have two structure to compare and overlapping them is a mess. But the dual views raise some user interface difficulties. Models are either displayed or undisplayed — there is no notion that a model could be displayed in one view but not in a second view. So how would the user control what is displayed in each view? Maybe the second view is special and it displays just the models it was told to display when it was created no matter whether they are shown or not. Also mouse mode input like selecting atoms by clicking them should work in both views. Again this will be some trouble because the second view is showing models that are considered hidden (thus not shown in the standard first view), and hidden models are not selectable with the mouse. Would image save include both views? That would be extra complication too.

[goddard@…]

Hi Elaine,

  I made a feature request for the split view idea.  The basic difficulty is if some of the views decide not to show some of the models then various code will break like selecting atoms, saving images, view all, popup atom balloons, … will all assume those displayed models are shown, unless the concept that not all displayed models are shown in all views is introduced throughout the code.  This is like the “lenses” we had in Chimera 1.  It is possible to do, but may not be worth the complexity.

	Tom

[Tom Goddard]

Begin forwarded message:

From: Elaine Meng
Subject: Re: [ChimeraX] #440: Dual view windows to compare structures
Date: October 4, 2016 at 11:38:55 AM PDT
To: Tom Goddard

As I understand it, this issue of having multiple different displays is completely different than the split/combined toggle that I thought Markus was getting at. I found the latter extremely useful in SYBYL, and in that case there is no issue of difference of displays. It is simply whether the models are in parallel universes or the same universe.

Elaine

[mheller@…]

Maybe not, Elaine.  Would you be able to explain that toggle?

-----Original Message-----
From: ChimeraX [mailto:ChimeraX-bugs-admin@cgl.ucsf.edu] 
Sent: Tuesday, October 04, 2016 12:42 PM
Subject: Re: [ChimeraX] #440: Dual view windows to compare structures

#440: Dual view windows to compare structures
----------------------------------+---------------------
          Reporter:  mheller@…    |      Owner:  goddard
              Type:  enhancement  |     Status:  new
          Priority:  major        |  Milestone:
         Component:  Graphics     |    Version:
        Resolution:               |   Keywords:
        Blocked By:               |   Blocking:
Notify when closed:               |   Platform:  all
           Project:  ChimeraX     |
----------------------------------+---------------------

Comment (by goddard):

 Begin forwarded message:

 From: Elaine Meng
 Subject: Re: [ChimeraX] #440: Dual view windows to compare structures
 Date: October 4, 2016 at 11:38:55 AM PDT
 To: Tom Goddard

 As I understand it, this issue of having multiple different displays is  completely different than the split/combined toggle that I thought Markus  was getting at.  I found the latter extremely useful in SYBYL, and in that  case there is no issue of difference of displays.  It is simply whether  the models are in parallel universes or the same universe.

 Elaine

--
Ticket URL: <https://plato.cgl.ucsf.edu/trac/ChimeraX/ticket/440#comment:5>
ChimeraX <http://www.rbvi.ucsf.edu/chimerax/>
ChimeraX Issue Tracker

[Elaine Meng]

Hi Markus,
What I was imagining was like a feature I’ve used before in SYBYL.  One can simply toggle the models between separated and together.  So if you had superimposed the structures in the unified view, in the split view they would be separated but comparable visually and would rotate individually.  Each structure only has one display (i.e. it can’t be colored differently or have different atoms shown in the split view versus the combined view).   In SYBYL the toggle was between one graphics window containing all the models, or four quadrants with four models separate and four centers of rotation (although really the same center as in the single unified view).  It was exact and did not suffer from the cumulative loss of synchronization as in Chimera where the separate centers of rotation are not equivalent to the unified center of rotation.  Although SYBYL had quadrants, it could be generalized to different tilings like halves, sixths, etc.

This imagined feature was the basis of my suggestion on chimera-users of how approximate it in Chimera.
Elaine

[goddard@…]

I probably didn’t make clear what I thought was problematic about split view.  There are two ways to do it.  One is that you move the models so they are side by side and then rotate each (keeping exactly synchronized).  The other is that you don’t move the models at all, instead there are two “cameras” and each one only sees one of the models.  I was talking about this latter approach in my previous messages.  It has some advantages — for instance, if I am using split view and ask for the RMSD of the superposition I get a sensible value, whereas if the models were moved to be side by side I get a very large wrong value.  Basically the advantage of the second approach is that the models remain their correct superimposed positions and the split views only change which model is displayed.

[mheller@…]

Elaine,

Thanks for the explanation.  This sounds *exactly* like what I'm looking for, and what VIDA offers!  I re-read my original post, and I now see that my word usage was poor.  Apologies for that!

Markus

-----Original Message-----
From: ChimeraX [mailto:ChimeraX-bugs-admin@cgl.ucsf.edu] 
Sent: Wednesday, October 05, 2016 10:36 AM
Subject: Re: [ChimeraX] #440: Dual view windows to compare structures

#440: Dual view windows to compare structures
----------------------------------+---------------------
          Reporter:  mheller@…    |      Owner:  goddard
              Type:  enhancement  |     Status:  new
          Priority:  major        |  Milestone:
         Component:  Graphics     |    Version:
        Resolution:               |   Keywords:
        Blocked By:               |   Blocking:
Notify when closed:               |   Platform:  all
           Project:  ChimeraX     |
----------------------------------+---------------------

Comment (by meng):

 {{{
 Hi Markus,
 What I was imagining was like a feature I’ve used before in SYBYL.  One  can simply toggle the models between separated and together.  So if you  had superimposed the structures in the unified view, in the split view  they would be separated but comparable visually and would rotate  individually.  Each structure only has one display (i.e. it can’t be  colored differently or have different atoms shown in the split view versus
 the combined view).   In SYBYL the toggle was between one graphics window
 containing all the models, or four quadrants with four models separate and  four centers of rotation (although really the same center as in the single  unified view).  It was exact and did not suffer from the cumulative loss  of synchronization as in Chimera where the separate centers of rotation  are not equivalent to the unified center of rotation.  Although SYBYL had  quadrants, it could be generalized to different tilings like halves,  sixths, etc.

 This imagined feature was the basis of my suggestion on chimera-users of  how approximate it in Chimera.
 Elaine
 }}}

--
Ticket URL: <https://plato.cgl.ucsf.edu/trac/ChimeraX/ticket/440#comment:7>
ChimeraX <http://www.rbvi.ucsf.edu/chimerax/>
ChimeraX Issue Tracker