Opened 3 weeks ago
Last modified 3 days ago
#19367 assigned defect
Replace RCSB ligand-expo URL which is being retired.
| Reported by: | Tom Goddard | Owned by: | Greg Couch |
|---|---|---|---|
| Priority: | moderate | Milestone: | 1.11 |
| Component: | Input/Output | Version: | |
| Keywords: | Cc: | brinda.vallat@…, ezra.peisach@…, Eric Pettersen, tcroll@… | |
| Blocked By: | Blocking: | ||
| Notify when closed: | Platform: | all | |
| Project: | ChimeraX |
Description
The Ligand Expo RCSB site is being retired and the following URL will cease to work at some point
The new URL to fetch a ligand .cif file is
ChimeraX is using this in the mmcif reader in mmcif.pyx _get_template(). This should be fixed for the ChimeraX 1.11 release.
This was reported by Brinda Vallat and Ezra Peisach:
On 11/13/25 9:07 AM, Brinda Vallat wrote:
Hi Tom,
Ezra (cc'd here) has pointed out that there are some references to ligand expo in ChimeraX.
We want to inform you that ligand expo is being shutdown. Please check to make sure that any old references are taken care of.
Please let us know if you have any questions.
Thank you,
Brinda
--
BRINDA VALLAT, Ph.D.
RCSB PDB
From: Ezra Peisach
Subject: Re: Ligand-expo to be shutdown
Date: November 13, 2025 at 6:47:58 AM PST
To: Brinda Vallat, Tom Goddard
For further details..
I noticed that you have a shared library that references ligand-expo. The centos-7 version of ChimeraX had a python module which performed downloads from there
% strings ./usr/libexec/UCSF-ChimeraX/lib/python3.11/site-packages/chimerax/mmcif/mmcif.cpython-311-x86_64-linux-gnu.so | grep ligand-ex
http://ligand-expo.rcsb.org/
If you are still downloading from ligand-expo, I suggest a changed the the RCSB PDB download API such as:
https://files.rcsb.org/ligands/download/HEM.cif
Thank you very much,
Ezra Peisach
Change History (9)
comment:1 by , 3 weeks ago
comment:2 by , 3 weeks ago
More examples:
http://ligand-expo.rcsb.org/reports/S/SER_LSN3_DHG/SER_LSN3_DHG.cif
http://ligand-expo.rcsb.org/reports/T/THR_LL_DHG1/THR_LL_DHG1.cif
http://ligand-expo.rcsb.org/reports/S/SER_LEO2_DHG/SER_LEO2_DHG.cif
http://ligand-expo.rcsb.org/reports/T/THR_LSN3_DHG1/THR_LSN3_DHG1.cif
http://ligand-expo.rcsb.org/reports/S/SER_LFZW_DHG/SER_LFZW_DHG.cif
http://ligand-expo.rcsb.org/reports/T/THR_LEO2_DHG1/THR_LEO2_DHG1.cif
http://ligand-expo.rcsb.org/reports/T/THR_LFZW_DHG1/THR_LFZW_DHG1.cif
comment:3 by , 3 weeks ago
I see. These are from the variant dictionary. I was not aware that ligand-expo was providing an interface to access this data. Let me get back to you and see if we can come up with a solution. --Ezra On 11/13/25 8:09 PM, ChimeraX wrote: > > >
comment:4 by , 3 weeks ago
| Cc: | added |
|---|
comment:5 by , 2 weeks ago
This shouldn't be a problem for ISOLDE... in response to some prior issues I've had the aa-variants-v1.cif (along with a "core_components.cif" containing the most common standard residues) bundled into the distribution for a while now. They're used by ISOLDE to fill in the blanks on truncated residues to get them simulation-ready... kept getting bug reports from people whose firewall settings were blocking the RCSB. From #19367 it looks like I'll need to update the bundled file, but otherwise everything should be fine. On Fri, Nov 14, 2025 at 7:59 PM ChimeraX <ChimeraX-bugs-admin@cgl.ucsf.edu> wrote: > > >
comment:6 by , 2 weeks ago
Hi Ezra,
You should be able to use the HTTP User-Agent in your ligand-expo web server logs to see which versions of ChimeraX are still using it. The user agent should look like 'UCSF-ChimeraX/1.11 (Windows 10.0.19045 AMD64)' -- so you can extract the ChimeraX version, the OS, and the CPU architecture.
It would be especially interesting to if there are any requests from recent versions of ChimeraX (so 1.9, 1.10, or 1.10.1), and if so, what entries they are getting.
comment:7 by , 3 days ago
Hi Ezra and Brinda,
Do you have an update for us? We're about to release a new version of ChimeraX in a few weeks and would like to make any needed changes before the release. The following release of ChimeraX will probably be in June 2026.
-- Greg
comment:8 by , 3 days ago
I think the best option is to cache/distribute the alternatives file. The only update in the last 15 years was one of dictionary compliance - there has not been a material change to the files that I am aware of. -Ezra On 12/1/25 5:38 PM, ChimeraX wrote: > > >
comment:9 by , 3 days ago
An advantage of using ligand-expo is that the rarely used variant templates don't add to the ChimeraX's huge download size. That is also a reason for not including components.cif (and the fact that components.cif is ever increasing in size).
Why not put the individual templates on wwpdb like the other residue templates? Either in https://files.wwpdb.org/pub/pdb/refdata/chem_comp/ or maybe a new subdirectory like https://files.wwpdb.org/pub/pdb/refdata/chem_comp_variants/.
Did you ever survey how much of a problem this really is with recent versions of ChimeraX? ChimeraX provides a distinctive user agent in its requests that includes the ChimeraX version number.
The problem is that https://files.rcsb.org/ligands/download/ doesn't have all of the ligands that are in ligand-expo. For example, http://ligand-expo.rcsb.org/reports/S/SER_LL_DHG/SER_LL_DHG.cif used to work. And https://files.rcsb.org/ligands/download/SER_LL_DHG.cif doesn't.
ChimeraX's current protocol for using ligand-expo is if the ligand name has an underscore in it. Otherwise it uses https://files.wwpdb.org/pub/pdb/refdata/chem_comp/{name[-1]}/{name}/{name}.cif.
This functionality is very important for the ISOLDE extension for ChimeraX.