[gmx-users] Error in free energy calculation (Chris Neale)

[Sadaf Rani]

Dear Chris
Thank you very much for giving your comprehensive information on my mdp
options.
In my current simulation, I am trying to switch off coulombic interactions
while keeping VDW on.
In gromacs manual, it is mentioned that with parallel simulations and/or
non-bonded force calculation on the GPU, a value of 20 or 40
often gives the best performance so I chose nstlist as 20. however, I am
not very much sure about the best value.
Simulations from lambda window 11 to 20 complete very well without any
lincs warning. I am getting following performance with the current
selection of mdp:-

          Core t (s)   Wall t (s)        (%)
       Time:   630938.079    39433.648     1600.0
                         10h57:13
                 (ns/day)    (hour/ns)
Performance:        6.573        3.651
Finished mdrun on rank 0 Mon Mar 23 07:41:56 2020



 I am running simulations from Lambda 0-1 in 21 different windows. For
Lambda 9 I am getting the following error:-

      Step           Time
        1100000     2200.00000

   Energies (kJ/mol)
           Bond Restraint Pot.          Angle    Proper Dih.  Improper Dih.
    6.18949e+03    2.44298e+00    1.68046e+04    2.06705e+04    9.49832e+02
          LJ-14     Coulomb-14        LJ (SR)  Disper. corr.   Coulomb (SR)
    7.91681e+03    7.10046e+04    2.30248e+05    0.00000e+00   -2.05702e+06
   Coul. recip.    Angle Rest.     Dih. Rest.      Potential    Kinetic En.
    8.96080e+03    7.31848e-01    7.99950e-01   -1.69427e+06    3.28779e+05
   Total Energy    Temperature Pres. DC (bar) Pressure (bar)      dVcoul/dl
   -1.36549e+06    3.00055e+02   -0.00000e+00   -1.54667e+01    2.05257e+03
       dVvdw/dl dVrestraint/dl   Constr. rmsd
   -3.75472e+02    3.97478e+00    3.30377e-06


-------------------------------------------------------
Program:     gmx mdrun, version 2020-UNCHECKED
Source file: src/gromacs/ewald/pme_redistribute.cpp (line 304)
MPI rank:    5 (out of 16)

Fatal error:
2 particles communicated to PME rank 5 are more than 2/3 times the cut-off
out
of the domain decomposition cell of their charge group in dimension y.
This usually means that your system is not well equilibrated.

For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors

I have equilibrated the system well and pressure is as below:-

Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Pressure                    1.65391       0.58     81.606    -2.0618  (bar)

How should I fix this error? Any suggestions for improvement will really
help me.
Many thanks.
Sadaf