Terminal -COOH naming issue with propka3 and CHARMM36m #196
Description
Hi all -- I cannot be the only one with this issue.
I am working with the CHARMM36m forcefield in GROMACS 2025.2 (note that this issue was also present in CHARMM27); when inputting a peptide with a CYS C-terminus, propka continuously spits out the following error: Unexpected number (7) of atoms in residue CYS 10 _ in conformation 1A. It seems that the program does not sufficiently account for potential variation in the naming of the oxygen atoms in carboxylic acids.
The parsing method used in the reading of input .pdb files, as found here in the source code, only seems to account for OXT as the possible three-character terminal oxygen indicator. In the case of CHARMM, and likely other forcefields, the standard naming convention for terminal carboxylic acids is OT1, OT2 or OT1, OT2, and HT2, depending on protonation. This will always raise an issue with propka, and the only way to bypass this error is to manually rename atoms in the .pdb file, which ultimately causes further GROMACS-related issues.
tl;dr: propka's parsing framework for C-terminal -COOH groups causes fatal atom counting issues.