COD bond-perception bug

[Tristan Croll]

Sorry for the probable double-report, but the ChimeraX Help/Report a Bug
didn't echo back to my email... if you do "open 7703621 from cod" you get a
bit of a mess, as pictured... apparently, for whatever reason, the standard
for metal-organic framework structures is to only explicitly report the
metal coordination bonds, not the covalent ones. ChimeraX's automatic bond
perception chokes on this - it skips any atom that already has a bond
specified, assuming that means *all* its bonds have been specified.
[image: image.png]

-- 
Altos Labs UK Limited | England | Company reg 13484917  
Registered 
address: 3rd Floor 1 Ashley Road, Altrincham, Cheshire, United Kingdom, 
WA14 2DT

[Tristan Croll]

Added by email2trac

[Tristan Croll]

Also, possibly worth tackling: this sort of lazily thrown together
structure is surprisingly common (hydrogens are actually "correct", just in
the wrong ASUs). There are symmetry-specifying bond records in the corecif
spec, but (of course) they're not always used... This particular example is
COD 7203220.
[image: image.png]

On Mon, Jul 13, 2026 at 5:07 PM Tristan Croll <tcroll@altoslabs.com> wrote:

>

[Tristan Croll]

Added by email2trac

[Tristan Croll]

Added by email2trac

[Greg Couch]

Duplicate of #20746. Likely fixed by recent change.

[Greg Couch]

FYI, today, unit cell support for small-molecule CIF files was added. But that doesn't reread the CIF files to find the cross-symmetry bonds. Let me know how you expect COD 7203220 to work.

[Tristan Croll]

Will give it a try when I get the chance to do an updated install. FWIW,
this isn't really for my benefit - I have everything I need working in
Clipper, and probably won't be spending much more time in small-molecule
world once this current work is done. But here's its working code for
reassembling COD 7203220 back into a sensible ASU (puts all the hydrogens
back with their parent phenols, and bonds them up):
https://github.com/tristanic/chimerax-clipper/blob/3bc1bfa676b08848ba484bbef0cd0f6de708f884/src/io/small_molecule.py#L296.
Feel free to use as you see fit - the ChimeraX-Clipper bundle is LGPL, and
I can't really claim to have written this code anyway.

On Wed, Jul 15, 2026 at 1:12 AM ChimeraX <ChimeraX-bugs-admin@cgl.ucsf.edu>
wrote:

>
>