Dear Elaine, Thanks a lot for your extensive and informative answer! I’ll send future questions to the mailing list. As you already assumed it’s a bit more complicated. I am in the final stages of refinement in ISOLDE. To perform some additinal simulations I probably cannot just edit the pdb file. However, your idea to use a different chain ID, this could work. I’ll check it out tomorrow. Many thanks again for your help. Kind regards, Mark
On 4 Feb 2025, at 19:29, Elaine Meng <meng@cgl.ucsf.edu> wrote:
On Feb 4, 2025, at 3:30 AM, Mark Hilge <markhilge@icloud.com> wrote:
Dear Dr. Meng,
[... details removed ...]
Question: In ChimeraX, is there a possibility to introduce a proteolytic cleavage between say Leu112 and R113? This should result in Leu112 with a carboxylate group followed by a TER keyword.
Many thanks in advance! Kind regards, Mark
Dear Mark, It depends what you want to use this for. If it's just for a figure, that is totally different than if you were going to do some extensive workflow afterwards.
If you have an input PDB file, one "dumb" way is to text-edit it to put the TER card in before opening in ChimeraX. However, that will not add a carboxylate oxygen, there would just be a dangling C=O. If you wanted an OXT you would have to add it "manually," e.g. in the Build Structure tool. However, that will be problematic because it won't want to add the OXT right on top of the N, which is where it would want to go. You'd also have to change the conformation of the chain, e.g. rotate a bond with the mouse mode or the "torsion" command. <https://www.rbvi.ucsf.edu/chimerax/docs/user/tools/buildstructure.html> <https://www.rbvi.ucsf.edu/chimerax/docs/user/commands/torsion.html>
Within ChimeraX, you could use command ~bond to remove bonds. However, if you do this within a chain defined in the input (e.g. PDB SEQRES lines or mmCIF file equivalent), it will then draw a dashed line because it still "knows" they are supposed to be connected within the chain because of the SEQRES records in the PDB file. <https://www.rbvi.ucsf.edu/chimerax/docs/user/commands/bond.html>
So it's all more complicated than you might think, especially if you wanted everything to be correct for some subsequent modeling workflow. You might instead give the two parts two different chain IDs. One way is to text-edit the PDB file. Another is to open the thing twice, delete the first part of the chain in one copy, delete the other part of the chain in the other copy, and then use "combine" to create a new model that has both parts but with different chain IDs, and then use "save" to save a PDB file of that combined model. <https://www.rbvi.ucsf.edu/chimerax/docs/user/commands/save.html#pdb>
However, you would still need to text-edit the PDB file manually so that there are SEQRES records for the two different chains (the two parts of the original single chain) that each contain only the residues for the corresponding part. These are different than the ATOM records that have the coordinates. And I can't even begin to explain how you might do it in mmCIF, all of this is in PDB format only. So if you bothered to do all that, then if you use "addh" it will automatically add the "OXT" because it will finally recognize the new cleaved C-terminus as a real C-terminus. <https://www.rbvi.ucsf.edu/chimerax/docs/user/commands/addh.html>
For future reference: please use chimerax-users@cgl.ucsf.edu <mailto:chimerax-users@cgl.ucsf.edu> address CC'd here to ask questions about ChimeraX, omitting any information that needs to remain private. If you send email to me only, it may not get answered.
Thanks, Elaine ----- Elaine C. Meng, Ph.D. UCSF Chimera(X) team Resource for Biocomputing, Visualization, and Informatics Department of Pharmaceutical Chemistry University of California, San Francisco