Tailoring Boltz-2 ChimeraX Output for AlphaBridge server Circos Plot Submission
Dear ChimeraX Team, I am currently working on visualizing interface residues using Circos plots via the Alphabridge web server, and I’m exploring ways to tailor the Boltz-2 ChimeraX output folder to serve as a compatible input format for Alphabridge web server (https://alpha-bridge.eu/) similar to the AF-3 output zipped folder. AlphaBridge<https://alpha-bridge.eu/> Zip file containing the AlphaFold resultSubmit job alpha-bridge.eu Despite successfully generating Boltz-2 outputs in ChimeraX daily build, I’m facing challenges in aligning the output folder with what Alphabridge web server expects for Circos plot generation. Could you kindly advise on: * Whether there is a recommended way to export or reformat Boltz-2 ChimeraX outputs for Alphabridge webserver compatibility? * Any scripts or workflows that facilitate this integration? Any advice on post-output work flow steps would be immensely helpful, as I aim to produce publication-quality Circos plot visualizations of interface residue networks similar to those demonstrated by recent AlphaBridge pre-print publication (DOI: https://doi.org/10.1101/2024.10.23.619601). Thank you very much for your support and for the fantastic tool you continue to develop for structural biology community. I look forward to your recommendations. Best regards, Arun Arun Gupta PhD, MRSC Sr. Post Doctoral Research Assistant Gillespie/McMichael Group Nuffield Dept. of Clinical Medicine, University of Oxford, Centre for Immuno-Oncology Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ Tel: + 44 (0)1865 2612913
Hi Arun, As you have found out Boltz 2 and AlphaFold 3 organize their input and output files in different directories than AlphaFold 3. Since Alphabridge is designed to work with AlphaFold 3 and not Boltz it is not surprising that AlphaBridge won't work with Boltz. But it is not just the arrangement of files that is the problem. The actual data output by Boltz aside from the mmCIF structure file uses different formats. Boltz for instance puts its PAE and PDE per-residue-pair confidence data in a numpy .npz file while AlphaFold 3 uses .json (at least last I used it). Also the confidence metrics like ipTM in Boltz are in a json file that is different from what AF3 provides. The pLDDT scores are in the b-factor column (and also an mmCIF scoring table I think) in both AF3 and Boltz 2 output so Alphabridge would be able to access those. In short, it would take more than just rearranging files, some would probably have to be put in different formats to work with Alphabridge. Maybe someone has already done that, you could ask the Alphabridge developers. ChimeraX could do it if we had time to implement it. But I'd be more inclined to just add directly into ChimeraX the Alphabridge plots so you don't need to go to a web server. Could you tell me which specific plots you like from Alphabridge? Is it the circular residue-residue interaction plot that colors by different interfaces that you like? (Example image below.) Tom 
On Aug 8, 2025, at 3:00 AM, Arun Gupta via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:
Dear ChimeraX Team, I am currently working on visualizing interface residues using Circos plots via the Alphabridge web server, and I’m exploring ways to tailor the Boltz-2 ChimeraX output folder to serve as a compatible input format for Alphabridge web server (https://alpha-bridge.eu/) similar to the AF-3 output zipped folder. AlphaBridge <https://alpha-bridge.eu/> Zip file containing the AlphaFold resultSubmit job alpha-bridge.eu <http://alpha-bridge.eu/> Despite successfully generating Boltz-2 outputs in ChimeraX daily build, I’m facing challenges in aligning the output folder with what Alphabridge web server expects for Circos plot generation. Could you kindly advise on: Whether there is a recommended way to export or reformat Boltz-2 ChimeraX outputs for Alphabridge webserver compatibility? Any scripts or workflows that facilitate this integration?
Any advice on post-output work flow steps would be immensely helpful, as I aim to produce publication-quality Circos plot visualizations of interface residue networks similar to those demonstrated by recent AlphaBridge pre-print publication (DOI:https://doi.org/10.1101/2024.10.23.619601). Thank you very much for your support and for the fantastic tool you continue to develop for structural biology community. I look forward to your recommendations. Best regards, Arun
Arun Gupta PhD, MRSC Sr. Post Doctoral Research Assistant Gillespie/McMichael Group Nuffield Dept. of Clinical Medicine, University of Oxford, Centre for Immuno-Oncology Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ Tel: + 44 (0)1865 2612913
_______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu <mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
Dear Tom, Thank you very much for your detailed reply and raising a feature request / ticket for the plugin integration of AlphaBridge’s plotting capabilities directly within ChimeraX — specifically, support for generating circular residue–residue interaction plots (Circos plots) that use colour to distinguish different interaction interfaces. I am particularly interested in analysing and making publication-quality figures for multi-interface protein complexes such as TCR–peptide–MHC protein complexes, where such Circos-style visualizations are highly informative in summarizing and comparing interfacial residues across multiple interaction surfaces between various protein complexes. Having these visual options natively in ChimeraX — as an AlphaBridge plots plugin — would significantly streamline workflows and make the comparative analysis of complex interfaces far more accessible. For context, I have already been in touch with the AlphaBridge team with the same request, and I am currently waiting for their response regarding the feasibility of taking Boltz-2 output as a compatible input format for AlphaBridge web server. Looking forward to the possibility of an early integration of the AlphaBridge plugin within ChimeraX, as this plugin feature would greatly enhance protein complex interface analysis workflows. Thank you again for considering this feature request. Best regards, Arun Arun Gupta PhD, MRSC Sr. Post Doctoral Research Assistant Gillespie/McMichael Group Nuffield Dept. of Clinical Medicine, University of Oxford, Centre for Immuno-Oncology Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ Tel: + 44 (0)1865 2612913 ________________________________ From: Tom Goddard <goddard@sonic.net> Sent: Friday, August 8, 2025 7:57 PM To: Arun Gupta <arun.gupta@immonc.ox.ac.uk> Cc: chimerax-users@cgl.ucsf.edu <chimerax-users@cgl.ucsf.edu> Subject: Re: [chimerax-users] Tailoring Boltz-2 ChimeraX Output for AlphaBridge server Circos Plot Submission Hi Arun, As you have found out Boltz 2 and AlphaFold 3 organize their input and output files in different directories than AlphaFold 3. Since Alphabridge is designed to work with AlphaFold 3 and not Boltz it is not surprising that AlphaBridge won't work with Boltz. But it is not just the arrangement of files that is the problem. The actual data output by Boltz aside from the mmCIF structure file uses different formats. Boltz for instance puts its PAE and PDE per-residue-pair confidence data in a numpy .npz file while AlphaFold 3 uses .json (at least last I used it). Also the confidence metrics like ipTM in Boltz are in a json file that is different from what AF3 provides. The pLDDT scores are in the b-factor column (and also an mmCIF scoring table I think) in both AF3 and Boltz 2 output so Alphabridge would be able to access those. In short, it would take more than just rearranging files, some would probably have to be put in different formats to work with Alphabridge. Maybe someone has already done that, you could ask the Alphabridge developers. ChimeraX could do it if we had time to implement it. But I'd be more inclined to just add directly into ChimeraX the Alphabridge plots so you don't need to go to a web server. Could you tell me which specific plots you like from Alphabridge? Is it the circular residue-residue interaction plot that colors by different interfaces that you like? (Example image below.) Tom [alphabridge.png] On Aug 8, 2025, at 3:00 AM, Arun Gupta via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote: Dear ChimeraX Team, I am currently working on visualizing interface residues using Circos plots via the Alphabridge web server, and I’m exploring ways to tailor the Boltz-2 ChimeraX output folder to serve as a compatible input format for Alphabridge web server (https://alpha-bridge.eu/) similar to the AF-3 output zipped folder. AlphaBridge<https://alpha-bridge.eu/> Zip file containing the AlphaFold resultSubmit job alpha-bridge.eu<http://alpha-bridge.eu/> Despite successfully generating Boltz-2 outputs in ChimeraX daily build, I’m facing challenges in aligning the output folder with what Alphabridge web server expects for Circos plot generation. Could you kindly advise on: * Whether there is a recommended way to export or reformat Boltz-2 ChimeraX outputs for Alphabridge webserver compatibility? * Any scripts or workflows that facilitate this integration? Any advice on post-output work flow steps would be immensely helpful, as I aim to produce publication-quality Circos plot visualizations of interface residue networks similar to those demonstrated by recent AlphaBridge pre-print publication (DOI:https://doi.org/10.1101/2024.10.23.619601). Thank you very much for your support and for the fantastic tool you continue to develop for structural biology community. I look forward to your recommendations. Best regards, Arun Arun Gupta PhD, MRSC Sr. Post Doctoral Research Assistant Gillespie/McMichael Group Nuffield Dept. of Clinical Medicine, University of Oxford, Centre for Immuno-Oncology Old Road Campus Research Building, Roosevelt Drive, Oxford OX3 7DQ Tel: + 44 (0)1865 2612913 _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu<mailto:chimerax-users@cgl.ucsf.edu> To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu<mailto:chimerax-users-leave@cgl.ucsf.edu> Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/
participants (2)
-
Arun Gupta -
Tom Goddard