[gmx-users] ERROR: non-zero virtual-site masses of atomtype MW* in TRP and TYR

[Guillermo Pérez-Hernández]

Hello!

So,

pdb2gmx -f 3CAP.pdb -ff amber99sb-ildn -water tip3p  -vsite aromatic

creates a topol.top that contains MW1 and MW2 atomtypes for TYR and TRP
residues[1].

However, when running
grompp -f md.mdp -c conf.gro -p topol.top

different things happen depending on the gromacs version[2]:

v. 4.6.5 : no problem
v. 5.0.2: ERROR
v. 5.1.1: ERROR

The error messages are (2 samples out of 28 times):
ERROR 1 [file topol.top, line 53291]:
  virtual site MW1 (Res TYR-10) has non-zero mass 0.504 (state A) / 0.504
  (state B)
       Check your topology.
[...]
ERROR 4 [file topol.top, line 53291]:
  virtual site MW1 (Res TRP-35) has non-zero mass 52.056 (state A) / 52.056
  (state B)
       Check your topology.

Any help would be appreciated. Of course, if this is a RTFM-situation,
please just point me towards the parts of the manual that I might have
overlooked. See the notes for details.

Best, and thanks in advance,

Guillermo

Notes:
[1] These atomtypes are in the center of the aromatic rings, and, as per
the manual "...each with a mass equal to the total mass of the
respective ring..."
[2] These versions are the ones available to me right now, there's no
other logic in the choice
[3] All three versions generate identical topol.top files. All three use
the amber99sb-ildn.ff folder that is shipped with each of them.
[4] The md.mdp file is a minimal file, down-edited from this sample
http://manual.gromacs.org/online/mdp.html
[5] The 3CAP.pdb is a "clean" pdb which runs cleanly through pdb2gmx in
all three versions



Dr. Guillermo Pérez-Hernández
Freie Universität Berlin
Institute for Mathematics
Arnimallee 6
D-14195 Berlin
tel       0049 30 838 75775
http://userpage.fu-berlin.de/gph82/
--