[gmx-users] Martini Coarse Graining Segmentation Fault using g_fg2cg

[Emanuel Peter]

Dear Gromacs Users,

I have performed the Martini Coarse Graining procedure for the
simulation of two protein domains by using the two scripts atom2cg.awk
and seq2itp.pl. Then I was able to perform my simulation of my coarse
grained system. My problem is now to convert my final cg structure into
an all atom structure. For this purpose I tried to use the tool g_fg2cg
with the inputs containing my fine grained topology ,the coarse grained
topology and my coarse grained structure file. But obviously all
attempts to apply that tool ended up in a segmentation fault message. I
also tried to convert my all atom structure into a cg structure which
ended up with the same segmentation fault message. Could you give me
some advice on that problem? Here is my screen output:
 Option     Filename  Type         Description
------------------------------------------------------------
-pfg LOV1_LOV2.pdb.top  Input        Topology file
-pcg martini_v2.0_example.top  Input, Opt!  Topology file
  -c lov1lov2_42ns_27.1.pdb  Input        Generic trajectory: xtc trr
trj gro
                                   g96 pdb
  -o        out.gro  Output       Generic structure: gro g96 pdb xml

      Option   Type  Value  Description
------------------------------------------------------
      -[no]h   bool     no  Print help info and quit
       -nice    int      0  Set the nicelevel
          -n    int      0  1: fg2cg transformation, 0: cg2fg
transformation
        -wat    int      0  1: rewrites FG_CG water to true fg water;
        -rad   real    0.3  A radius for random atom insertion;

calling cpp...
processing topology...
Generated 279 of the 1225 non-bonded parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 3 bonded neighbours for Protein_B 1
NOTE:
  System has non-zero total charge: -2.999998e+00

#   G96BONDS:   2738
#  G96ANGLES:   3994
#      PDIHS:   1465
#      IDIHS:   1328
#       LJ14:   4520
calling cpp...
processing topology...
Generated 0 of the 465 non-bonded parameter combinations
Excluding 1 bonded neighbours for Protein 1
Excluding 1 bonded neighbours for ProteinB 1
NOTE:
  System has non-zero total charge: -6.000000e+00


Number of fg atoms 2696
Number of cg atoms 571
Reading frame       0 time 42000.004   1264673691
Segmentation fault

When I applied this tool on the ubiquitin example I had another problem,
but I did it in the same way as before. Here is the screen output:

Option     Filename  Type         Description
------------------------------------------------------------
-pfg       1ubq.top  Input        Topology file
-pcg       out3.top  Input, Opt!  Topology file
  -c       1UBQ.gro  Input        Generic trajectory: xtc trr trj gro
g96 pdb
  -o        out.gro  Output       Generic structure: gro g96 pdb xml

      Option   Type  Value  Description
------------------------------------------------------
      -[no]h   bool     no  Print help info and quit
       -nice    int      0  Set the nicelevel
          -n    int      1  1: fg2cg transformation, 0: cg2fg
transformation
        -wat    int      0  1: rewrites FG_CG water to true fg water;
        -rad   real    0.3  A radius for random atom insertion;

calling cpp...
processing topology...
Generated 279 of the 1225 non-bonded parameter combinations
Excluding 3 bonded neighbours for Protein_A 1
Excluding 2 bonded neighbours for SOL 58
#   G96BONDS:   766
#  G96ANGLES:   1107
#      PDIHS:   428
#      IDIHS:   334
#       LJ14:   1304
#     SETTLE:   58
calling cpp...
cpp: out3.top: Datei oder Verzeichnis nicht gefunden
cpp: warning: '-x c' after last input file has no effect
cpp: no input files
cpp exit code: 256
Tried to execute: 'cpp 
-I/pc50417/pee18323/REVERSE_trans_gromacs/agromacs-reverse/share/top 
out3.top > grompp605Sk3'
The 'cpp' command is defined in the .mdp file
processing topology...

Number of fg atoms 934
Number of cg atoms 0
Reading frames from gro file 'UBIQUITIN', 934 atoms.
Reading frame       0 time    0.000   out.xtc
Last frame          0 time    0.000

Back Off! I just backed up out.gro to ./#out.gro.2#

Cg structure computed !

In that case out.gro did not contain any atoms.

Thanks for any suggestions!

Best regards,

Emanuel Peter