[gmx-users] Re: Ligand-protein interaction scoring

[Douglas Houston]

Hi,

I have noticed something I didn't expect; when I use nsteps=0, near the
end of md.log shows 31 steps (for my test case):

           Step           Time         Lambda
            31       31.00000        0.00000

Stepsize too small, or no change in energy.
Converged to machine precision,

When I use nstep=1 OR set emtol to some huge number like 10000, then I
indeed only get 1 step. Perhaps nstep=0 is interpreted as "continue
running for as many steps as are required to reach convergence"?

My .mdp file looks like:

cpp               =  /lib/cpp
define            =  -DFLEX_SPC
constraints       =  none
integrator        =  steep
nsteps            =  0
nstxout           =  5000
nstvout           =  5000
nstlog            =  5000
nstenergy         =  250
nstxtcout         =  250
nstlist           =  10
emtol             =  1
emstep            =  0.01
nstcomm           =  0
rlist             =  1
rcoulomb          =  1.0
rvdw              =  1.0
Tcoupl            =  no
Pcoupl            =  no
gen_vel           =  no
pbc               =  no

Thanks,

Doug