Simulations fail instantly ("Sim termination reason: None") across entire model, all fidelity levels, survives full restart + reinstall
Hi all, I'm running into a persistent issue in ISOLDE and would appreciate any guidance, especially from Tristan if he has a moment. ChimeraX version: 1.12 OS: Windows Summary: isolde sim start fails immediately with no simulation actually running, for essentially any selection in my model -- not just one problem region. The log shows: Sim termination reason: None ISOLDE: stopped sim with no further error detail, immediately after the residue templates (FAD, P6G, PO4) finish loading from the internal database. What I've ruled out so far: * Not a stuck/lingering Unparameterised Residue widget (confirmed via full ChimeraX restart) * Not specific to one problem residue -- originally isolated a specific residue (ASN, chain A) via bisection, but after fixing it (full atom+bond rebuild from a working homologous chain, confirmed via distance checks that backbone geometry was reasonable), the same failure then appeared in a completely unrelated, previously-working region of the model * Not a duplicate-atom/clash issue -- scanned for overlapping atoms (<0.6 A) in the affected region, found and removed a set of stray duplicate hydrogens, confirmed zero overlaps remaining; simulation still failed * Not an mmCIF metadata issue -- the same region on the model's other (homodimer) chain simulates without any problem * Not a session-state/cache artifact -- fails identically after closing ChimeraX entirely and reopening a freshly saved PDB from disk * Not a simulation fidelity/resource issue -- tested at Medium/Medium and Lowest/Fastest, identical failure both times * Not resolved by a full OS-level restart (not just closing ChimeraX) -- identical failure immediately after reboot, on a select /A:60-61 range that had previously worked earlier the same day * Not resolved by reinstalling the ISOLDE bundle via the ChimeraX Toolshed and restarting -- identical failure persists on previously-failing selections (/A:508-518, /A:60-86) immediately after reinstall; only the one range that had always worked (/A:60-61) "worked," which isn't diagnostic since it never failed in the first place Sequence that reproduces it (after a clean restart): open session_backup_03.cxs isolde start clipper associate #1.1.1.1 toModel #1.2 select /A:508-518 isolde sim start /A:508-518 Result: templates load (FAD/P6G/PO4), then immediate "Sim termination reason: None." The Problem Zones tab is empty, which I believe is expected since no simulation has ever successfully started to generate restraint data. Model contains a protein homodimer (two chains, otherwise identical) plus FAD, PO4, and P6G heteroatoms, refined via Servalcat from a cryo-EM map. Happy to share the model/map files if useful for reproducing this. Any guidance on what else to check (OpenMM installation, platform/GPU selection, log verbosity settings for more detail than "None") would be much appreciated John Guardado (he/him) | G1 PhD Student, BM CALIFORNIA INSTITUTE OF TECHNOLOGY
This sounds odd. Any chance you can share the offending .cxs off-list so I can try to reproduce the problem? On Mon, 6 Jul 2026 at 19:32, Guardado, John H. via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hi all, I'm running into a persistent issue in ISOLDE and would appreciate any guidance, especially from Tristan if he has a moment. ChimeraX version: 1.12 OS: Windows Summary: isolde sim start fails immediately with no simulation actually running, for essentially any selection in my model -- not just one problem region. The log shows: Sim termination reason: None ISOLDE: stopped sim with no further error detail, immediately after the residue templates (FAD, P6G, PO4) finish loading from the internal database. What I've ruled out so far:
- Not a stuck/lingering Unparameterised Residue widget (confirmed via full ChimeraX restart) - Not specific to one problem residue -- originally isolated a specific residue (ASN, chain A) via bisection, but after fixing it (full atom+bond rebuild from a working homologous chain, confirmed via distance checks that backbone geometry was reasonable), the same failure then appeared in a completely unrelated, previously-working region of the model - Not a duplicate-atom/clash issue -- scanned for overlapping atoms (<0.6 A) in the affected region, found and removed a set of stray duplicate hydrogens, confirmed zero overlaps remaining; simulation still failed - Not an mmCIF metadata issue -- the same region on the model's other (homodimer) chain simulates without any problem - Not a session-state/cache artifact -- fails identically after closing ChimeraX entirely and reopening a freshly saved PDB from disk - Not a simulation fidelity/resource issue -- tested at Medium/Medium and Lowest/Fastest, identical failure both times - Not resolved by a full OS-level restart (not just closing ChimeraX) -- identical failure immediately after reboot, on a select /A:60-61 range that had previously worked earlier the same day - Not resolved by reinstalling the ISOLDE bundle via the ChimeraX Toolshed and restarting -- identical failure persists on previously-failing selections (/A:508-518, /A:60-86) immediately after reinstall; only the one range that had always worked (/A:60-61) "worked," which isn't diagnostic since it never failed in the first place
Sequence that reproduces it (after a clean restart): open session_backup_03.cxs isolde start clipper associate #1.1.1.1 toModel #1.2 select /A:508-518 isolde sim start /A:508-518 Result: templates load (FAD/P6G/PO4), then immediate "Sim termination reason: None." The Problem Zones tab is empty, which I believe is expected since no simulation has ever successfully started to generate restraint data. Model contains a protein homodimer (two chains, otherwise identical) plus FAD, PO4, and P6G heteroatoms, refined via Servalcat from a cryo-EM map. Happy to share the model/map files if useful for reproducing this. Any guidance on what else to check (OpenMM installation, platform/GPU selection, log verbosity settings for more detail than "None") would be much appreciated
*John Guardado *(he/him) | G1 PhD Student, BM
CALIFORNIA INSTITUTE OF TECHNOLOGY _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
-- Altos Labs UK Limited | England | Company reg 13484917 Registered address: 3rd Floor 1 Ashley Road, Altrincham, Cheshire, United Kingdom, WA14 2DT
Could you also report this via Help/Report a Bug after triggering it? That will provide the full log contents, which might be helpful in diagnosis. On Mon, Jul 6, 2026 at 7:39 PM Tristan Croll <tcroll@altoslabs.com> wrote:
This sounds odd. Any chance you can share the offending .cxs off-list so I can try to reproduce the problem?
On Mon, 6 Jul 2026 at 19:32, Guardado, John H. via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hi all, I'm running into a persistent issue in ISOLDE and would appreciate any guidance, especially from Tristan if he has a moment. ChimeraX version: 1.12 OS: Windows Summary: isolde sim start fails immediately with no simulation actually running, for essentially any selection in my model -- not just one problem region. The log shows: Sim termination reason: None ISOLDE: stopped sim with no further error detail, immediately after the residue templates (FAD, P6G, PO4) finish loading from the internal database. What I've ruled out so far:
- Not a stuck/lingering Unparameterised Residue widget (confirmed via full ChimeraX restart) - Not specific to one problem residue -- originally isolated a specific residue (ASN, chain A) via bisection, but after fixing it (full atom+bond rebuild from a working homologous chain, confirmed via distance checks that backbone geometry was reasonable), the same failure then appeared in a completely unrelated, previously-working region of the model - Not a duplicate-atom/clash issue -- scanned for overlapping atoms (<0.6 A) in the affected region, found and removed a set of stray duplicate hydrogens, confirmed zero overlaps remaining; simulation still failed - Not an mmCIF metadata issue -- the same region on the model's other (homodimer) chain simulates without any problem - Not a session-state/cache artifact -- fails identically after closing ChimeraX entirely and reopening a freshly saved PDB from disk - Not a simulation fidelity/resource issue -- tested at Medium/Medium and Lowest/Fastest, identical failure both times - Not resolved by a full OS-level restart (not just closing ChimeraX) -- identical failure immediately after reboot, on a select /A:60-61 range that had previously worked earlier the same day - Not resolved by reinstalling the ISOLDE bundle via the ChimeraX Toolshed and restarting -- identical failure persists on previously-failing selections (/A:508-518, /A:60-86) immediately after reinstall; only the one range that had always worked (/A:60-61) "worked," which isn't diagnostic since it never failed in the first place
Sequence that reproduces it (after a clean restart): open session_backup_03.cxs isolde start clipper associate #1.1.1.1 toModel #1.2 select /A:508-518 isolde sim start /A:508-518 Result: templates load (FAD/P6G/PO4), then immediate "Sim termination reason: None." The Problem Zones tab is empty, which I believe is expected since no simulation has ever successfully started to generate restraint data. Model contains a protein homodimer (two chains, otherwise identical) plus FAD, PO4, and P6G heteroatoms, refined via Servalcat from a cryo-EM map. Happy to share the model/map files if useful for reproducing this. Any guidance on what else to check (OpenMM installation, platform/GPU selection, log verbosity settings for more detail than "None") would be much appreciated
*John Guardado *(he/him) | G1 PhD Student, BM
CALIFORNIA INSTITUTE OF TECHNOLOGY _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
-- Altos Labs UK Limited | England | Company reg 13484917 Registered address: 3rd Floor 1 Ashley Road, Altrincham, Cheshire, United Kingdom, WA14 2DT
participants (2)
-
Guardado, John H. -
Tristan Croll