[gmx-users] Energy minimization

[Andrea Bacconi]

Hi All,
I'm continuing my MD simulation in water, everything is ok up to the 
energy minimization step. I think I understood the problem but I don't 
know how to solve it.
When I run EM in void all is ok Fmax is around 25 but when I add water 
I'm not able to minimize the system's energy (only Fmax >>1000) using 
steep descent alghorithm.
Now, when I use cg, instead I get this error message from grompp : 
ERROR: can not do Conjugate Gradients with constraints (59565)
My em.mdp file is the following.

cpp                      = /lib/cpp
constraints            =none
define                  = -DFLEX_SPC
integrator             = cg
dt                        = 0.002
nsteps                  = 100000
nslist                    = 10
ns_type               = grid
rlist                      = 0.9
coulombtype        = PME
rcoulomb              = 1.8
rvdw                     = 0.9

nstcgsteep            = 100
emtol                    = 1000.0
emstep                  = 0.001

I looked up in the mailing list for a similar problem, I found it but it 
was not usefull. To my mind it looks like that the program sees some 
consraints when water is around because when I simulate the same protein 
in void everything is fine sd and cg (no problem at all).
So please help me if you can
thank you for your attention
Andrea