Query regarding Dihedral restrain

[saikat0003]

GROMACS version:2024.2
GROMACS modification: No
1KCal/mol rad2 dihedral restrain works in the AMBER package. However, the same dihedral restrain (1KCal/mol rad2) does not work for the gromacs package. I have used the same forcefield (amberff14sb) and water model (tip4pew) for both the cases (the systems are identical). When I used 2KCal/mol rad2 dihedral restrain it works fine.

[MagnusL]

First, I hope you are aware that GROMACS dihedral restraints are specified in kJ/(mol rad²) (see File formats - GROMACS 2024.2 documentation). So, if you want the restraint potentials to be the same you could try 4.184 kJ/(mol rad²).

What do you mean with 1 kcal/(mol rad²) (presumably 1 kJ/(mol rad²)) does not work, but 2 kcal/(mol rad²) works? Does the simulation crash? Or is the restraint potential not strong enough? (See above.)

I don’t think there are any official amber14sb ports in GROMACS yet. Are you sure that the force field is correctly converted?

[saikat0003]

Thanks for the reply. I used 4.184 kJ/(mol rad²) dihedral restrain in gromacs. My question is, in the AMBER forcefield equation, the Vn is actually Vn/2 . In gromacs is it the same
in the forcefield equation?