Add a command to set the sequence for an atomic model chain

[matthias.vorlaender@…]

Want a way to set the full sequence of chains in an atomic model, even for residues that have no coordinates so they can be written into mmCIF and PDB files for depositing in the PDB database.

From: Vorländer,Matthias Kopano via ChimeraX-users <chimerax-users@…>
Subject: [chimerax-users] Re: How to save a sequence in an mmCIF file?
Date: December 21, 2023 at 5:33:38 AM PST
To: ChimeraX Users Help <chimerax-users@…>
Reply-To: Vorländer,Matthias Kopano

Hi everyone,

I also just struggled with the demands of CIF file deposition to the PDB for a ~50 chain RNA-protein complex and the time it takes to copy-paste sequence into different web interfaces. Would it be possible to interface the ChimeraX sequence association functionalities to write the associated sequence into the mmCIF file? At the end of building large models using many cryo-EM maps the structures often have been cut apart and pieced together many times, so my dream scenario would be to use “alpafold match #model trim false“ to directly write the full-length sequence of the alphafold model into the mmcif file or PDB header for the matches chain. Would be of huge value for PDB deposition!

Many thanks and best wishes,
Matthias

[Elaine Meng]

Maybe I misunderstood, but I thought the suggestion was to have a "save" option for mmCIF that would include the sequence associated with each structure chain in ChimeraX (using our existing structure-sequence association mechanism, presumably after opening a sequence file, or showing the sequence from another structure chain that already had complete SEQRES, or perhaps the easiest, doing a uniprot sequence fetch). That way the user would not need to specify the sequence in some separate command or command option.

[Eric Pettersen]

No, I believe the idea is to copy sequence information from one chain to another, or from an alignment association. That allows the information to be saved in either PDB or mmCIF formats.

Anyway, this ticket is basically a duplicate of #10270, so I will be closing this one and adding Matthias to the cc list for the other ticket.