Hi Alexis, The likely problem with the ligand is that ChimeraX is guesstimating atom hybridization states from bond lengths, bond angles, and other related criteria, and sometimes it gets the hybridization estimate wrong, which results in an incorrect protonation. The reason that hydrogen addition (and particularly H-bond-guided addition) only works on whole models it that it is done in multiple passes, and later passes rely on having hydrogens from earlier passes present and correct, and therefore it would be really complicated to only protonate part of a model correctly. Probably the only feasible method for partially protonating a model would be to remember the initial protonation state of the model, add hydrogens, then remove any hydrogens that were added outside the target area. I think this is would be doable. I'll open a ticket for it. For now, I'd say your best approach is to open a second copy of the model, protonate it, delete the non-waters, delete the waters from the other model and then combine the two. --Eric Eric Pettersen UCSF Computer Graphics Lab

On Mar 30, 2026, at 7:33 PM, Alexis Rohou via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Hi all,

As specified in the documentation, in command addh, "Hydrogens are added to entire atomic model(s) even if only parts of models are specified."

Given how specifically this is stated, I'm guessing there was a good rationale for doing it this way.

However I am facing a situation where I'd really, really like to add hydrogens only to waters (which were added by Servalcat). For reasons that I haven't got to the bottom of, addh adds hydrogens in the wrong places in a ligand I'm working on (a peptide with a mixture of natural and non-canonical amino acids, though I'm almost certain I've observed this behavior with other, less exotic, ligands).

So: what is/was the rationale for not allowing users to add hydrogens to only parts of a model? How would the developers feel about a pull request adding an option to override this behavior and only add hydrogens to the spec supplied by the user?

Cheers, Alexis _______________________________________________ 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/