[Chimera-users] Color sub-region with marker

Tom Goddard goddard at sonic.net
Wed Nov 3 16:09:04 PDT 2010

Hi Xiaofeng,

   The measure contactarea command can only handle surfaces consisting 
of a single surface pieces.  Surfaces can have more than one selectable 
piece, but molmap makes a single piece surface so I'm puzzled by the 
error you get.  I see that if you have an outline box shown for the 
molmap surface then that is considered a second surface piece and will 
generate the error message you see.

   For your second problem, I can put markers on the clipping surface 
cap using "place markers on surface" but the trouble is the markers are 
clipped so you don't see them.  The near/far clipping planes clip 
everything.  You do see half of a spherical marker (clipped in half) if 
you hide the map after placing the marker on the clip surface.  I'd 
suggest using Per-Model Clipping (menu Tools / Depiction) to clip just 
the map surface -- that way you will be able to see the markers you 
place on the clip plane.


> Hi Tom,
> Thanks for your reply. The questions are:
> (1) When I follow your procedure until "measure contactarea #0 #1 12 color tan", it promotes "#0 has 2 surface pieces, require1" and stop working. I don't understand what that means(figure attached).
> (2) The question of number 3 at last time about how to put markers directly on the clipped plane is that, for example, in Side view window, when I move the front bar (normally the front and behind bar just cover the object) into the object to observe inside, I try to put markers on this capped plane inside the object, but it doesn't work even I have chosen "put markers on surface".
> Xiaofeng
> ----- Original Message -----
> Hi Xiaofeng,
>     The Chimera Color Zone tool has lots of limitations as you've noticed.
>     The jagged edges on the colored region when using color zone happen
> because the surface is made up of triangles and Chimera can only color
> each triangle vertex and then linearly interpolates colors across the
> triangle based on the 3 vertex values.  That can look bad.  Your
> solution of using the "subdivide surface" option in the volume dialog
> surface surface and mesh options panel will help reduce the jaggedness.
> There is one tricky solution that can produce a smoother edge.  It uses
> a command I just added a month ago called "measure contactarea".  The
> virtue of this approach is that it creates a new surface chopping the
> triangles at the edge of your zone to make a perfect smooth boundary.
> But it wasn't intended as a substitute for the color zone dialog.
> Instead it measures the area on one surface within a given distance of
> another surface and displays that contact area.  The first problem using
> this is that it doesn't use markers and a surface like color zone --
> instead it uses two surfaces.  So the first trick is to turn your
> markers into a surface.  If the markers are model #1 use
>       molmap #1 3
> to make a 3 Angstrom resolution surfaces for those markers.  Now to
> color a zone around those markers on volume data #0 use
>       measure contactarea #0 #1 12 color tan
> This will make a new surface matching volume surface #0 within a range
> of 12 Angstroms from surface #1 (the marker surface created with
> molmap).  It has beautiful smooth edges.  See the attached picture.
> You'll need a Chimera daily build to use "measure contactarea".  I
> suggest tonight's daily build, because I made some changes to "measure
> contactarea" to make the surface look very smooth.  Earlier daily builds
> had a less smooth edge, and less smooth colored region.  If you add
> "slab 2" as an option it will make the colored patch have thickness 2A
> (not shown).
>     1) You can't use different color zone radii on the same surface at
> the same time.  This is item 37 in the Chimera feature request list.
>       http://plato.cgl.ucsf.edu/trac/chimera/wiki/requests
> You can use the above trick multiple times to create color patches using
> different radii.
>     2) Not sure I understand this question.  I think you are asking if
> you can color inside rectangles and ellipses.  There's no way to do that
> now.  I'm not sure how you would describe the rectangle or ellipse with
> a marker since a marker is a sphere and those shapes have non-equivalent
> axes.  You can create such shapes with the Chimera "shape" command but
> you can't for instance color everything within such a shape.  The
> closest you can do is color near a traced path made with markers and
> links.  This basically lets you color in a cylindrical zone.  To enable
> this you use the bondzone command
>       http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/bondzone.html
>     3) Not sure exactly what case you are interested in.  If you clip a
> surface you can place markers exactly of the "cap" surface that covers
> the hole left by clipping.  You would use the Volume Tracer mouse mode
> with "Place markers on surfaces" enabled.  If you wanted to place
> markers on single data planes of a volume you would enable "Place
> markers on data planes".
>       Tom
>> Dear Chimera users,
>> I always have problems when I color a sub-region of density map by markers. Rather than a zigzag, boundary, I try to create a smooth curve boundary of sub-region from a whole density 3D volume. In this case, if I just put one marker and select a big coloring radius, the boundary in cross-view will show a circle curve at the boundary, however if I use a very small coloring radius and put more and closer markers on the boundary, the boundary shows a sawtooth in cross-view, even I choose 4 times for subdividing surface. Or is there any other way to do this instead of markers? And I have several other questions:
>> (1) Is that possible to have different coloring radius in one Chimera session?
>> (2) except for ball shape, can we have more options on the coloring object with markers? For example rectangular, ellipse and so on.
>> (3) Is that possible to put markers directly on the clip planes?
>> Thanks in advance,
>> Xiaofeng

More information about the Chimera-users mailing list