[gmx-users] Simulated Tempering Issues
Alexander Yang
ahy3nz at virginia.edu
Fri Jan 6 18:45:06 CET 2017
Dear Gromacs users,
I've been attempting simulated tempering (ST) in Gromacs to study gel-phase
bilayers (~15000 atoms), but I've been encountering some difficulties. I've
been following similar steps in the expanded ensemble tutorial on
alchemistry.org and adapting them for ST. The system I am studying is
already fairly well-equilibrated, so I don't imagine any huge movements in
the system.
I've picked very modest temperature values (305 K - 350 K with 2.5 K
spacing), yet I've been getting domain decomposition errors upon restarting
the simulation. I've done brief simulations (same MD parameters) on
clusters for 15 minutes and 30 minutes without crashing. When continuing
either simulation, the system crashes and throws the domain decomp error.
On the other hand, running these simulations indefinitely on the head node
does not exhibit crashes at the steps that have crashed for restarts after
15 minutes. I'm not sure if this is a coincidence with a ST move error, an
error with my MD parameters, or something with restarts. Could someone
advise? I've copied the error, last few lines of the log, and mdp below.
----Error----
Fatal error:
An atom moved too far between two domain decomposition steps
This usually means that your system is not well equilibrated
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
------------
-----Log---
DD step 3854999 vol min/aver 0.401 load imb.: force 11.2% pme
mesh/force 2.059
Step Time Lambda
3855000 7710.00000 0.00000
Energies (kJ/mol)
G96Bond G96Angle Proper Dih. Improper Dih. LJ (SR)
1.31304e+04 1.57687e+04 9.63926e+03 7.34404e+02 -9.18268e+03
Disper. corr. Coulomb (SR) Coul. recip. Potential Kinetic En.
-2.70660e+03 -2.50854e+05 1.12235e+03 -2.22348e+05 5.38027e+04
Total Energy Temperature Pres. DC (bar) Pressure (bar)
-1.68545e+05 3.58572e+02 -1.70205e+02 -2.93216e+02
DD step 3859999 vol min/aver 0.391 load imb.: force 7.7% pme
mesh/force 2.344
Step Time Lambda
3860000 7720.00000 0.00000
MC-lambda information
Wang-Landau incrementor is: 1
N Temp.(K) Count G(in kT) dG(in kT)
1 305.000 193 0.00000 193.00000
2 307.500 0 193.00000 0.00000
3 310.000 0 193.00000 0.00000
4 312.500 0 193.00000 0.00000
5 315.000 0 193.00000 0.00000
6 317.500 0 193.00000 0.00000
7 320.000 0 193.00000 0.00000
8 322.500 0 193.00000 0.00000
9 325.000 0 193.00000 -35.00000
10 327.500 35 158.00000 -96.00000
11 330.000 131 62.00000 -95.00000
12 332.500 226 -33.00000 -95.00000
13 335.000 321 -128.00000 -95.00000
14 337.500 416 -223.00000 -95.00000
15 340.000 511 -318.00000 -95.00000
16 342.500 606 -413.00000 -97.00000
17 345.000 703 -510.00000 -93.00000
18 347.500 796 -603.00000 -91.00000 <<
19 350.000 887 -694.00000 0.00000
Energies (kJ/mol)
G96Bond G96Angle Proper Dih. Improper Dih. LJ (SR)
1.24610e+04 1.60933e+04 9.86938e+03 7.51926e+02 -8.56113e+03
Disper. corr. Coulomb (SR) Coul. recip. Potential Kinetic En.
-2.69740e+03 -2.52430e+05 1.08555e+03 -2.23427e+05 5.48813e+04
Total Energy Temperature Pres. DC (bar) Pressure (bar)
-1.68546e+05 3.65761e+02 -1.69051e+02 3.42231e+02
DD step 3864999 vol min/aver 0.397 load imb.: force 9.8% pme
mesh/force 1.822
Step Time Lambda
3865000 7730.00000 0.00000
Energies (kJ/mol)
G96Bond G96Angle Proper Dih. Improper Dih. LJ (SR)
1.33244e+04 1.56493e+04 9.63355e+03 7.38294e+02 -7.52212e+03
Disper. corr. Coulomb (SR) Coul. recip. Potential Kinetic En.
-2.67952e+03 -2.52827e+05 1.16878e+03 -2.22514e+05 5.50394e+04
Total Energy Temperature Pres. DC (bar) Pressure (bar)
-1.67475e+05 3.66815e+02 -1.66819e+02 5.74291e+02
+04
Total Energy Temperature Pres. DC (bar) Pressure (bar)
-1.71349e+05 3.54322e+02 -1.63023e+02 -7.17865e+00
----
----MDP------
;title = Simulated Tempering Run
; Run parameters
integrator = md-vv
nsteps = 25000000 ; 50ns
dt = 0.002
; Output control
nstxout = 0 ; Don't save coordinates
nstvout = 0 ; Don't save velocities
nstenergy = 5000
nstlog = 5000
nstxtcout = 5000
;bond parameters
continuation = yes
constraint_algorithm = lincs ;lincs
constraints = hbonds
lincs_iter = 1
lincs_order = 4
; Neighbor searching
cutoff-scheme = Verlet
ns_type = grid
nstlist = 10
rcoulomb = 1.4
rvdw = 1.4
;Electrostatics
coulombtype = PME
pme_order = 4
fourierspacing = 0.16
; Temperature coupling
tcoupl = v-rescale ;v-rescale
tc-grps = non-water water
tau_t = 0.1 0.1
ref_t = 305 305
;Pressure coupling
pcoupl = berendsen;berendsen
pcoupltype = isotropic
tau_p = 1.0 ; ps
ref_p = 1.0 ; bar
compressibility = 4.5e-5
refcoord_scaling = com
;PBC
pbc = xyz
;Dispersion correction
DispCorr =EnerPres
;Velocity generation
gen_vel = no
;Simulated tempering
nstexpanded = 800
simulated-tempering = yes
sim-temp-low = 305
sim-temp-high = 355
simulated-tempering-scaling = linear
init_lambda_state = 0
temperature_lambdas = 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90
0.95 1.00
lmc-stats = wang-landau
lmc-move = metropolized-gibbs
lmc-weights-equil = wl-delta
weight-equil-wl-delta = 0.001
wl-scale = 0.7
wl-ratio = 0.8
init-wl-delta = 1.0
wl-oneovert = yes
-------------
Thanks for the help,
Alex
More information about the gromacs.org_gmx-users
mailing list