There are at least two issues with depositing mmCIF files written by ChimeraX. 1. As you found, if ChimeraX did not get an authoritative sequence from the input data, then it does not output the sequence. For PDB deposition you need to associate the complete sequence with a chain. A good way to do that in ChimeraX is to open the complete sequence in the Sequence Viewer dialog and right-click to use the Sequence / Update Chain Sequence dialog. On the command line, use the "sequence update" command. If your model has atomic coordinates for **everything** in the complete sequence, you can use the bestGuess mmCIF save option to put the sequence in the saved file. 2. ChimeraX does not keep track of sheets. To output complete secondary structure information in the mmCIF output, you have to use the computedSheets mmCIF save option. That runs a dssp computation to figure out the sheets and associated strands. The PDB has its own variation of the dssp program that it uses, so this might not be important. HTH, Greg On 11/6/2024 4:51 PM, Eric Pettersen via Chimera-users wrote:
Hi David, For PDB output, ChimeraX should output SEQRES records for each chain as long as either: (A) the chain had sequence information in the original input file (/e.g./ SEQRES records), or (B) the chain has no missing-structure pseudobonds. I not the author of the mmCIF I/O part of ChimeraX, but I believe the criteria is different in that the chain has to have its "full_sequence_known" attribute set to true. You can do this "by hand" with the command (assuming your structure is model #1) "setattr #1 chains full_sequence_known true".
--Eric
Eric Pettersen UCSF Computer Graphics Lab
P.S. There are different lists for Chimera vs. ChimeraX questions. I have cc'ed the ChimeraX list on this.
On Nov 6, 2024, at 10:54 AM, David Bhella via Chimera-users <chimera-users@cgl.ucsf.edu> wrote:
Dear Chimera/ChimeraX team, I have a model for an asymmetric unit of a virus capsid that has three chains of the same capsid protein – but I am having trouble depositing my model in the PDB because it only has a seqres entry for one of the three chains (B). Whatever I do, chimeraX will only write entries for that chain. Is there a way to force chimeraX (or any other program) to write the seqres card for each chain correctly? Thanks D *David Bhella* Professor of Structural Virology Director, Scottish Centre for Macromolecular Imaging Associate Director,MRC - University of Glasgow Centre for Virus Research
+44 (0)141 330 3685 <tel:+441413303685>
image001.png image002.png _______________________________________________ Chimera-users mailing list --chimera-users@cgl.ucsf.edu To unsubscribe send an email tochimera-users-leave@cgl.ucsf.edu Archives:https://mail.cgl.ucsf.edu/mailman/archives/list/chimera-users@cgl.ucsf.edu/
_______________________________________________ Chimera-users mailing list --chimera-users@cgl.ucsf.edu To unsubscribe send an email tochimera-users-leave@cgl.ucsf.edu Archives:https://mail.cgl.ucsf.edu/mailman/archives/list/chimera-users@cgl.ucsf.edu/
