FEP error: perturbed, excluded non-bonded pair interactions beyond the pair-list cut-off

User discussions
1

GROMACS version: 2024-2
GROMACS modification: No
We are trying to run a FEP simulation in which both A state and B state contain some dummy atoms. The topology file looks like this:
[ atoms ]
; nr type resi res atom cgnr charge mass ; qtot bond_type
1 c3 1 QNG C01 1 0.058400 12.01000 DMC 0.0000 12.01; qtot 0.058
2 DMC 1 QNG C02 2 0.000000 12.01000 ; qtot 0.600
3 DMC 1 QNG C03 3 0.000000 12.01000 ; qtot 0.384
4 DMC 1 QNG C04 4 0.000000 12.01000 ; qtot 0.307
5 ca 1 QNG C05 5 0.273700 12.01000 DMC 0.0000 12.01; qtot 0.696
However, we get the following error: Program: gmx mdrun, version 2024.2
Source file: src/gromacs/nbnxm/exclusionchecker.cpp (line 184)

Fatal error:
There are 7 perturbed, excluded non-bonded pair interactions beyond the
pair-list cut-off, which is not supported. This can happen because the system
is unstable or because intra-molecular interactions at long distances are
excluded. If the latter is the case, you can try to increase nstlist or rlist
to avoid this.The error is likely triggered by the use of couple-intramol=no
and the maximal distance in the decoupled molecule exceeding rlist."

Note that we are not using couple-moltype in my mdp file, therefore the couple-intramol is irrelevant. The FEP transformation is done by defining the A state and B state atoms in the tolopolgy as shown above and by setting the relevant lambda values in the mdp file, e.g.:
free_energy = yes
sc_alpha = 0.5
sc-power = 1
sc-sigma = 0.3
sc-coul = yes
nstdhdl = 100
init-lambda-state = 5
coul-lambdas = 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
vdw-lambdas = 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1.0

We are grateful for any suggestions on what went wrong here. Thanks!

Nanjie

2

The error must then be caused by atoms 1 and 5 being at a distance of more than the cut-off length. I assume that then non-bonded interactions between atoms 1 and 5 are excluded.

3

Hi, thanks for helping us figuring this out. Actually, the error will persist even when we increased the nonbond cut-off length to cover the entire molecule.
Also we find that the error has to do with having dummy atoms in the A state but not in the B state. So in this case, if we make the A state of the atoms 2, 3 and 4 to have regular atom types like c3 and ca (BTW, these are AMBER GAFF atom types), then there is no error. Once we bring dummy atom types like DMC into the A state of any atoms, then we get errors such as: “…There are 7 perturbed, excluded non-bonded pair interactions beyond the pair-list cut-off…”

We will try to create a small molecule complete case to reproduce the problem.
For now we are using a workaround to do what we want, avoiding putting dummy atoms in the A state.

Thanks again,
Nanjie

4

Are the dummy atoms bonded to each other and/or the normal atoms?

5

Yes, all the Dummy atoms are bonded together, separated by 1 to several bonds. We try to disappear two groups of atoms in a solute molecule (96 atoms) in two consecutive steps: in the first step, group 1 (42 atoms, all bonded together) is transformed into Dummy atoms, i.e. their B states are all Dummys. And in the second step, group 2 atoms are converted to Dummy atoms. In this step, the A-state of the group 1 atoms are Dummys. The first step runs OK, but the second step gives errors, most likely because somehow A-state cannot contain Dummys?

Thank you,
Nanjie