Transparency Command Mockup
Quick links and legend:
The following is basically equivalent to Chimera1
transparency.
Usage:
transparency
spec
percent
[ frames N ]
[ target string ]
The transparency command sets transparency
to the specified percent, where 0% is completely opaque
and 100% is completely transparent, without otherwise changing colors.
The frames option indicates making the change
gradually over N frames.
Still not settled where the spec should go.
Except for the frames option,
this mockup is analogous to that for single-color assignment
with the color command.
See also: color
The target option allows acting on only
a subset of the possible displays of a given atom, residue, etc.
The string can be any combination of the following letters,
without commas or spaces:
- a - atoms/bonds (halfbond coloring)
- l - atom labels and bond labels?
- s - surfaces
(molecular surfaces divisible to per-atom patches,
all surface models divisible to selected triangles)
- c - cartoons (divisible to per-residue segments)
- r - residue labels
- n - nonmolecular models (opacity component of model-level color)
- m - molecular models (opacity component of model-level color)
- b - bonds only
(halfbond mode off) with both end atoms specified
- p - pseudobonds with both end atoms specified
- d - distances and other pseudobond labels
Can bonds/pbonds be specified by selection of the bond/pbond alone?
The default target is all but the last three
(alscrnm), except further limited to
just atoms/bonds and their associated labels and surface patches (als)
when spec
contains only explicit specifications of atoms by atom name,
atom type, element, and/or atom attribute.
(These comments are essentially the same as for the color command)
Is this doable? By explicit specification
I mean that although higher levels could be included in spec,
all the “leaves” would be down to the atom level, for example:
:48-56@ca
#0@n,ca,c,o
/A:48@NZ/B:200@OD1
ligand z<5 & S
- I rather like including molecular model-level opacity in the default,
but could be dissuaded since it could be inefficient to change the model level
all the time when users are intending to act on specific atoms/residues.
- Should these designations be mutually exclusive?
(Does n include s, or only nonmolecular surfaces, or
no surfaces at all? Above,
I made them mutually exclusive in that n refers to the model level,
s to assignments at finer levels)
- Do we need
nucleotides,
ring fill,
pseudobond groups?
Ringfill and nucleotide assignments could be slaved to atom/bond assignments,
but I could imagine wanting ringfill-only transparency.
-
Could remove p,d and leave pseudobond coloring
(including the opacity component) to setattr
and/or pbond creation commands such as distance and findhbond.