NIH3D "Surface Hydrophobicity" preset not working

[kristen.browne@…]

As per summary

[pett]

Going to "sharpBoundaries false" broke coloring by hydrophobicity because it uses "color byattr" which in turn calls surface.color_atom_patches(), which seems to only work with sharp boundaries ("Surface #1.1 does not have atom patches, cannot color by atom"). I guess I don't understand why per-atom coloring is limited to sharp boundaries, but Tom must know why.

I have uploaded a new version (1.1.14) to the toolshed that uses sharp boundaries only for the hydrophobicity preset.

[Tom Goddard]

Associating each surface vertex with an atom that allows coloring the surface per-atom is only enabled if sharp boundaries are used. This was intentional. The per-atom patches calculated without sharp edges are random sawtooth patterns that are an artifact of the surface mesh. The reason that is not calculated by default is it looks horrible. See the attached image sharp_edges.png for a comparison. The sense of it is that if you only care about single color surfaces then you can use "sharp false" and it will speed up the computation and make a surface with substantially fewer triangles. If you want it multi-colored we want you to get a good quality appearance without very apparent artifacts. We could add an option to allow you to get the atom patches without sharp edges. But it is something I think no one would use. Which raises the question, why is NIH 3D trying to use it?

[Tom Goddard]

Comparison of per-atom surface coloring with and without sharp edges.

[pett]

I think sharp=false is just to reduce the vertex count of the surface, but if sharp=true is needed for a good appearance for a particular preset, I think that overrides the vertex-count priority.

[kristen.browne@…]

Agreed!

-----Original Message-----
From: ChimeraX <ChimeraX-bugs-admin@cgl.ucsf.edu>
Sent: Monday, October 2, 2023 5:56 PM
To: Browne, Kristen (NIH/NIAID) [C] <kristen.browne@nih.gov>; pett@cgl.ucsf.edu
Cc: goddard@cgl.ucsf.edu
Subject: [EXTERNAL] Re: [ChimeraX] #9878: NIH3D "Surface Hydrophobicity" preset not working

#9878: NIH3D "Surface Hydrophobicity" preset not working
---------------------------------------+--------------------
          Reporter:  kristen.browne@...  |      Owner:  pett
              Type:  defect            |     Status:  closed
          Priority:  moderate          |  Milestone:
         Component:  Depiction         |    Version:
        Resolution:  fixed             |   Keywords:
        Blocked By:                    |   Blocking:
Notify when closed:                    |   Platform:  all
           Project:  ChimeraX          |
---------------------------------------+--------------------
Comment (by pett):

 I think sharp=false is just to reduce the vertex count of the surface, but  if sharp=true is needed for a good appearance for a particular preset, I  think that overrides the vertex-count priority.
--
Ticket URL: <https://www.rbvi.ucsf.edu/trac/ChimeraX/ticket/9878#comment:3>
ChimeraX <https://www.rbvi.ucsf.edu/chimerax/>
ChimeraX Issue Tracker
CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and are confident the content is safe.