[gmx-users] energy minimization and normal mode analysis (again again probably)

[Tommy Carstensen]

Hi gmx-users,

I've tried to do pre-NMA energy minimization on 1CRN.pdb using the mdp
parameters listed at the bottom of this posting. Unfortunately the Fmax does
not converge to my emtol of 0.01 during the EM. I get the 'famous' error:
"Stepsize too small, or no change in energy. Converged to machine precision,
but not to the requested precision Fmax"
The values of Fmax, which I have attained, are all in the range 1-100. I
have tried different values of emstep with all combinations of
EM-integrators and columbtypes and it still does not work. What am I doing
wrong?

mdp parameters:
emtol           = 0.000001
integrator      = cg
xtc_grps        = Protein
constraints     = none
epsilon_r       = 1
title           = parameters_em
vdwtype         = cut-off
include         = -I../top
nsteps          = 50000
define          = -DFLEX_SPC
rlist           = 1 (tried with 0 as well)
emstep          = 0.1 (tried smaller and larger values)
nstcgsteep      = 100 (tried with default as well)
coulombtype     = PME (tried cut-off as well)
ns_type         = grid
nstlist         = 5 (tried with 0 well)
rcoulomb_switch = 0
nstxout         = 50
pbc             = xyz
rcoulomb        = 1
cpp             = /lib/cpp
energygrps      = Protein
rvdw-switch     = 0
rvdw            = 1

Please help or I might Go Really Offensive and Mess All Computer Systems.

Cheers,
Tommy Carstensen