Gromacs-CP2K interface discrepancy

User discussions
1

GROMACS version: 2024.3
GROMACS modification: No

Hello everyone,

I’m using GROMACS 2024.3 interfaced with CP2K 2024.3 to perform QM/MM simulations. After successfully running gmx grompp with the -qmi flag and starting the run with the generated .tpr file, I noticed an inconsistency in the input files automatically generated by GROMACS for CP2K.

Specifically:

  • The .pdb file generated for CP2K uses 0-based atom numbering.
  • The .inp file (particularly in the QMMM and LINK sections) uses 1-based atom indexing.

This mismatch leads to confusion and potentially incorrect QM region definitions, especially when defining QM_KIND or LINK atoms.
Is this expected behavior? If so, what is the recommended way to ensure consistency between the files?
If it’s a bug, I’m happy to help provide minimal working examples or further details.

Thanks in advance for any help!

Best regards,
Roberto