[Chimera-users] surfaces, electrostatic coloring
meng at cgl.ucsf.edu
Sat May 3 09:51:14 PDT 2008
On May 2, 2008, at 8:04 PM, James Fethiere wrote:
> Another wquestion. I managed to display separate surfaces for 2
> chains in a single model using the surfcat command you've
> recommended last time. Is this command accessible through the menu?
There is no menu equivalent to "surfcat," sorry.
> However, I can't color each surface separately according to
> electrostatics. I know that chimera does not calculate
> electrostatics. What I tried is to go to tools/surface binding/
> electrostatic surface coloring, set the palette to radius and set
> the full range. What I get when I color is a gradation from red to
> blue with no atom type discrimination.
When you choose "radius" the coloring is by distance from a point,
not electrostatics.. To actually color by electrostatics, you would
have to open an electrostatic potential map and then choose it from
that menu (the one with "radius" in it). Also, your two surfaces
would be two separate entries in the menu listing the surfaces. You
would have to choose one, color, and then choose the other, color again.
Perhaps it is misleading to call this tool "Electrostatic Surface
Coloring" when it is really identical to the "Surface Color" tool:
it includes options to color by distance from a point, axis, or
plane. The radial coloring is used (for example) to color a virus
capsid by the distance from its center to highlight the contours of
its surface; see the figure in the man page:
Another possibly confusing thing is that in older versions of
Chimera, after you use Surface Color, it can be tricky to change the
color of the surfaces. In current versions (recent daily builds),
you should just be able to color the surface with the menu
(Actions... Color) or commands (color, rangecolor) regardless of
whether this tool had been used.
> So I go to the attributes of this map in the model panel, and color
> source atoms. Great, I get red for acidic and blue for basics.
> Now for the other chain, it doesn't work. At the color source atoms
> step, the whole surface takes one single color. The other thing is
> that in the model panel I do not have the small color icon for the
> two maps!! Is there a better way to do this?
The two surfaces would again be listed separately in the Model Panel,
and have their own separate attributes dialogs. I don't know what
you mean by "doesn't work." Did you try to use the menu or commands
to color the surface? Also, I don't know what other things you did
in the session, but the red and blue you were seeing were probably
just the element colors: surfaces of oxygen atoms (red) and nitrogen
atoms (blue). Maybe you had colored the atoms of one chain and not
the other. In any case, you could proceed to color the surfaces with
the menu and/or commands as mentioned above.
If you don't have an electrostatic potential map and don't want to
color by distance from a point, axis, or plane, there is no reason to
use Surface Color. For example, to show the element colors on the
surface for the two chains in 2zcp:
surfcat one :.a & protein
surfcat two :.b & protein
col byatom (or "col byhet" to not change carbon color)
Or, instead of using the atomic element colors, you could
color :asp,glu residues red and :his,arg,lys residues blue.
> Can you also tell me how to make the surfaces smoother. Whatever
> vertex density I give, I see pretty rough edges and it's not very
The vertex density is the only way I know of for making the surface
smoother. I am surprised it is not doing that in your case - it
definitely works in the molecular surfaces I have tried: 5.0 is a
lot better, 10.0 even better. Did you press return? The main
drawback I ran into is that it can take a long time to calculate,
rotate, etc. Maybe increasing the probe radius would improve your
I hope this helps,
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Chimera-users