[gmx-users] substrate explodes

[YOLANDA SMALL]

Hello gmx-users,

I did an energy minimization on a protein system for which I constructed a
topology file for the unique substrate and cofactor at the active site.  I used
the freezegrp option to freeze the substrate and cofactor in all directions (my
.mdp file is below).  The energy minization does not converge but delivers this
message: 

Stepsize too small (7.49853e-07 nm)Converged to machine precision,
but not to the requested precision (1000)

Steepest Descents did not converge in 75 steps
  Potential Energy  =  3.08892e+08
Maximum force:  1.99058e+06

Upon inspection of the output structure, the substrate and cofactor are the only
things that have exploded even though I thought that freezing them keeps them
fixed.  Does anyone know what could be causing this problem?

mdp filecpp                 =  /lib/cpp
define              =  -DFLEX_SPC
; constraints         =  all-bonds
integrator          =  steep
nsteps              =  300
;
;       Energy minimizing stuff
;
emtol               =  1000
emstep              =  0.0001

nstcomm             =  1
ns_type             =  grid
pbc                 =  xyz
rlist               =  1
coulombtype         =  cut-off
rcoulomb            =  1.4
vdwtype             =  cut-off
rvdw                =  1.4
Tcoupl              =  no
Pcoupl              =  no
gen_vel             =  no
;
;      Frozen Groups
;
freezegrps          =  ORO FMN
freezedim           =  Y Y Y Y Y Y

Many thanks,
Yolanda