[gmx-users] restraint/freeze issues

[Ken Rotondi]

Hello all,

I'm having problems with position restraints. I'm trying test 
simulations of a tripeptide with the Calpha of residue 1 restrained 
with x,y,z constants of 10,000 each. When I run gmxdump on my .tpr file 
I find the following:

functype[100]=POSRES, pos0A=( 2.01600003e+00, 1.98000002e+00, 
0.00000000e+00), fcA=( 0.00000000e+00, 0.00000000e+00, 0.00000000e+00), 
pos0B=( 2.01600003e+00, 1.98000002e+00, 0.00000000e+00), fcB=( 
0.00000000e+00, 0.00000000e+00, 0.00000000e+00)
             functype[101]=POSRES, pos0A=( 1.60999998e-01, 
1.99699998e+00, 0.00000000e+00), fcA=( 1.00000000e+04, 1.00000000e+04, 
1.00000000e+04), pos0B=( 1.60999998e-01, 1.99699998e+00, 
0.00000000e+00), fcB=( 1.00000000e+04, 1.00000000e+04, 1.00000000e+04)
             functype[102]=POSRES, pos0A=( 1.99000001e-01, 
1.43000007e-01, 0.00000000e+00), fcA=( 0.00000000e+00, 0.00000000e+00, 
0.00000000e+00), pos0B=( 1.99000001e-01, 1.43000007e-01, 
0.00000000e+00), fcB=( 0.00000000e+00, 0.00000000e+00, 0.00000000e+00)
etc.

so yes, the appropriate atom is restrained. However, when I preform a 
50ps simulation at 400K the peptide just vibrates. Similar behavior was 
seen on a longer peptide simulated at 500K for 3ns, which prompted 
these tests. Upon looking through the gmxdump of my test .tpr file I 
found this at the end:

Group statistics
T-Coupling  :      26    627      1  (total 654 atoms)
Energy Mon. :      26    627      1  (total 654 atoms)
Acceleration:     654  (total 654 atoms)
Freeze      :     654  (total 654 atoms)
User1       :     654  (total 654 atoms)
User2       :     654  (total 654 atoms)
VCM         :     654  (total 654 atoms)
XTC         :      26    628  (total 654 atoms)
Or. Res. Fit:     654  (total 654 atoms)

Question 1) How do I group the CL- (1) with the SOL (627) in the same 
temperature bath, as Xavier has suggested?

Question 2) Does the "Freeze :   654  (total 654 atoms)" entry indicate 
that my system is indeed frozen?

Question 3) If the answer to question 2 is negative, what is a next 
logical step?

I include my .mdp file for a slightly longer test run. I see nothing 
unusual about it, but perhaps clearer eyes...

As always, many thanks,

Ken

;       .mdp file for NVT test MD of peptide fragments
;       0.2 ns
;	User Ken
;	Wed Nov 18 11:22:44 2004
;	Input file
;
title               =  Chicken
cpp                 =  /lib/cpp
define              = -DPOSRES
constraints         =  h-bonds
integrator          =  md
dt                  =  0.002	; ps !
nsteps              =  100000	; total 0.2 ns.
nstxout             =  5000     ; cord to .trr file every 10 ps.
nstvout             =  5000     ; velo to .trr file every 10 ps.
nstfout             =  5000	; force to .trr file every 10 ps.
nstlog              =  5000     ; energ to log file every 10ps.
nstenergy           =  250      ; energ to energ file every 0.5ps
nstxtcout           =  250      ; cood to .xtc file every 0.5ps
xtc_grps            =  Protein
nstlist             =  10       ; freq (steps) update neighbor list
ns_type             =  grid
pbc                 =  xyz      ; turns on periodic bound. cond.
rlist               =  0.95
coulombtype         =  switch
rcoulomb            =  0.9
rcoulomb_switch	    =  0.7
vdwtype             =  switch
rvdw                =  0.9
rvdw_switch         =  0.7
; Nose-Hover temperature coupling is on in three groups
tcoupl              =  nose-hoover
tc_grps		    =  Protein	SOL	CL-	
tau_t               =  0.1	0.1	0.1
ref_t               =  408      408	408
; Energy monitoring
energygrps          =  Protein  SOL	CL-
; Generate velocites is on at 408 K.
gen_vel             =  yes
gen_temp            =  408.0
gen_seed            =  -1