[gmx-users] pull code core dump

Daniel Santos simul at oninetspeed.pt
Tue Oct 12 12:48:57 CEST 2004


Dear all,

I am trying to do a free energy calculation by thermodynamic integration. My reaction coordinate is the distance between two centres of mass (molecule near a lipid bilayer and a cylinder containing part of the bilayer near the molecule). The distance between these two centres of mass is restrained during the simulation.

I can start the simulations with equilibrated systems and everything seams to go well for a while (the file pull.pdo containing information about the total forces felt by the molecule is created with success). Since at a random step Gromacs just core dumps, I'm thinking that maybe there is a bug in the code (I certainly hope that the problem is simpler)...

Some details:
 
 1) I generate the pull.ndx with:
  make_ndx -f box.gro -o pull.ndx
and remove the extra index numbers (I keep only the lipid and teste molecule 
atom indexes).
 
 2)My pull.ppa contains the following settings: 
 verbose                 = no
 runtype                 = constraint
 reftype                 = dynamic
 
 reference_group         = PLA
 group_1                 = PSO
 
 constraint_direction    = 0.0 0.0 1.0
 pulldim                 = N N Y
 reflag                  = 1
 
 r                       = 1.0
 rc                      = 1.5


Any help will be appreciated. Thanks in advance.

Daniel dos Santos





More information about the gromacs.org_gmx-users mailing list