[gmx-users] Warning: 1-4 interaction
Moeed
lecielll at googlemail.com
Fri Mar 11 17:41:49 CET 2011
Dear experts,
I have been trying to do NVT on a system (a single polyethylene chain in 110
hexane) in the past few days but still no success. For some reason
simulations are taking much longer than before. A run that used to complete
in a few hours is taking several weeks. I am wondering if there is something
wrong with our cluster or..my runs crash (LINCS) at early stages of
simulation.
step 100, will finish Wed Mar 16 00:08:53 2011
step 200, will finish Wed Mar 16 00:05:57 2011
step 300, will finish Wed Mar 16 00:04:05 2011
step 400, will finish Wed Mar 16 00:03:37 2011
Step 460, time 0.92 (ps) LINCS WARNING
relative constraint deviation after LINCS:
rms 22.298270, max 137.226395 (between atoms 2136 and 2138)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
2123 2124 92.6 0.1090 0.1866 0.1090
2139 2142 90.0 0.1090 9.1450 0.1090
2139 2141 90.1 0.1090 9.2416 0.1090
2139 2140 90.0 0.1090 9.1782 0.1090
***********************************************************
Here is more details:
To diagnose the reason for crash I watched the trajectory and noticed the
problem lies in hexane molecules and not the polymer. All LINCS and 1-4
warnings are because of hexane. So I followed the instruction on gmx site
and simplified the problem. Topology is not the root cause since a 2 ns run
on a single hexane works well. I am not getting any warnings before MD step
so be ignored by maxwarn either. Regarding bad intial structure I should say
that I was initially using emtol= 1000 and then I used 100 and the box is
too big and molecules are far apart. (box is 20nm containing 110 hexane)
I also did PR where no error message come up. As I move on to MD NVT systems
shows instabilities.
I dont know how else to overcome this situation which I have been working on
nearly one week.
pbc = xyz
;energygrps = PE HEX
; Run control
integrator = md
dt = 0.002
nsteps = 1000000 ;5000
nstcomm = 100
; Output control
nstenergy = 100
nstxout = 1
nstvout = 0
nstfout = 0
nstlog = 1000
nstxtcout = 1000
; Neighbor searching
nstlist = 10
ns_type = grid
; Electrostatics/VdW
coulombtype = PME
vdw-type = Shift
rcoulomb-switch = 0
rvdw-switch = 0.9 ;0
; Cut-offs
rlist = 1.25
rcoulomb = 1.25 ;1.1
rvdw = 1.0
; PME parameters
fourierspacing = 0.12
fourier_nx = 0
fourier_ny = 0
fourier_nz = 0
pme_order = 4
ewald_rtol = 1e-5
optimize_fft = yes
; Temperature coupling
Tcoupl = v-rescale
tc-grps = System ;HEX
tau_t = 0.1 ;0.1
ref_t = 300 ;300
; Pressure coupling
Pcoupl = no;Parrinello-Rahman
Pcoupltype = isotropic
tau_p = 1 1 ;0.5
compressibility = 4.5e-5 4.5e-5
ref_p = 20 20
; Velocity generation
gen_vel = yes
gen_temp = 300.0
gen_seed = 173529
; Bonds
constraints = all-bonds ;
constraint-algorithm = lincs
; Free energy control stuff
free_energy = no;yes
init_lambda = 0
delta_lambda = 0
sc_alpha = 0.5
sc-power = 1
sc_sigma = 0.3
;foreign_lambda = 0.1
dhdl_derivatives = yes
;couple-moltype = Polymer
couple-lambda0 = vdw-q ;vdw
couple-lambda1 = none ;vdw;none
couple-intramol = yes
nstdhdl = 10
separate_dhdl_file = yes
dh_hist_size = 0
dh_hist_spacing = 0.1
>
> Moeed wrote:
>
>> Hello Justin,
>>
>> I just scanned through the manual and also looked into the list to learn
>> about the significance of reciprocal part to realize your statement. I did
>> not find much about the fact that how I can judge if fourierspacing is the
>> cause of the crash (I mean relevance between fourierspacing and coulb
>> recip). I suspect the cause of crashes is because of
>>
>
> The manual will not describe every detail, but an understanding of the
> algorithms (through reading the appropriate literature) should provide you
> with what you need. Take Mark's advice and leave fourierspacing at its
> default value unless you are 100% sure of the effects of messing with it.
>
>
> electrostatics settings. BEfore that I had only vdw no run crashed.
>>
>
> That was a broken model, as we've discussed. Don't believe that its
> results are reliable.
>
>
> Since I used to get LINCS warning I removed bond constraints but still
>>
>
> LINCS warnings don't mean you should turn off constraints. With a 2-fs
> timestep and an all-atom model, you absolutely need to be using constraints.
>
> http://www.gromacs.org/Documentation/Errors#LINCS.2fSETTLE.2fSHAKE_warnings
>
>
> system is showing instabilities. I am confident that topology is fine and
>> just wondering how else one can inspect root cause for unphysical forces.
>>
>>
> The best advice I can give has already been given:
>
>
> http://www.gromacs.org/Documentation/Terminology/Blowing_Up#Diagnosing_an_Unstable_System
>
> Watch the trajectory. See what falls apart.
>
> You're also applying 70 bar of pressure to a system that may not yet be
> properly equilibrated for such conditions. I have found that the
> Parrinello-Rahman barostat is not particularly well-suited to harsh
> conditions, tight coupling, or simulations of any system that is far from
> equilibrium. You might be better off with Berendsen (and a fairly long
> tau_p) until your system stabilizes.
>
> -Justin
>
>
>> fourierspacing =0.12 : Coul. recip. 636.217 8.69433
>> 8.62908 0.0167398 3.68278
>>
>> fourierspacing =0.3 : Coul. recip. 642.921 6.06063
>> 6.02832 -0.00984063 -2.16496
>>
>>
>>
>>
>>
>> Everything has a cause; nothing is accidental. No stable run should
>> ever show LINCS warnings like this, so yes, there is something to fix.
>>
>> You also have not proven that your fourierspacing is not the cause.
>> The proper term to analyze is Coul-recip, which corresponds to
>> reciprocal space Coulombic term. I don't see that listed below.
>>
>> -Justin
>>
>>
>> Thanks,
>> Best,
>> Moeed
>>
>> ========================================
>> step 164300, remaining runtime: 241 s step 164400,
>> remaining runtime: 240 s step 164500, remaining
>> runtime: 240 s step 164600, remaining runtime: 239
>> s Warning: 1-4 interaction between 260 and 267 at
>> distance 73679654.990 which is larger than the 1-4 table size
>> 2.250 nm
>> These are ignored for the rest of the simulation
>> This usually means your system is exploding,
>> if not, you should increase table-extension in your mdp file
>> or with user tables increase the table size
>> =========================================
>> results till t=320 ps (before first run crash)
>> fourierspacing = 0.12
>>
>> LJ-14 757.246 26.6239 26.5661
>> -0.0189929 -6.07777
>>
>> Coulomb-14 -186.509 15.9866 15.9529
>> -0.0112174 -3.5896
>>
>> LJ (SR) -2691.84 117.026 107.231
>> -0.507373 -162.361
>>
>> Coulomb (SR) 725.662 17.6483 17.5876
>> 0.0158275 5.06483
>>
>> fourierspacing = 0.3
>>
>>
>> LJ-14 757.129 26.4936 26.4578
>> -0.0149128 -4.77214
>>
>> Coulomb-14 -199.523 14.949 14.158
>> -0.0519427 -16.6218
>>
>> LJ (SR) -2670.5 111.201 109.688
>> 0.19784 63.3092
>>
>> Coulomb (SR) 739.794 16.1807 15.4473
>> 0.0521352 16.6834
>>
>> ====================================================
>> pbc = xyz ; Run
>> control integrator = md
>> dt = 0.002 nsteps =
>> 1000000 ;5000 nstcomm = 100
>> ; Output control
>> nstenergy = 100 nstxout
>> = 100 nstvout = 0
>> nstfout = 0
>> nstlog = 1000 nstxtcout =
>> 1000 ; Neighbor searching
>> nstlist = 10 ns_type =
>> grid ; Electrostatics/VdW
>> coulombtype = PME vdw-type
>> = Shift rcoulomb-switch = 0
>> rvdw-switch = 0.9 ;0 ; Cut-offs
>> rlist = 1.25 rcoulomb
>> = 1.25 ;1.1 rvdw = 1.0
>> ; PME parameters
>> fourierspacing = 0.3 fourier_nx = 0
>> fourier_ny = 0
>> fourier_nz = 0
>> pme_order = 4 ewald_rtol = 1e-5
>> optimize_fft = yes
>>
>> ; Temperature coupling Tcoupl = v-rescale
>> tc-grps = System ;HEX
>> tau_t = 0.1 ;0.1 ref_t
>> = 400 ;300 ; Pressure coupling
>> Pcoupl = Parrinello-Rahman Pcoupltype
>> = isotropic tau_p = 1
>> 1 ;0.5 compressibility = 4.5e-5 4.5e-5
>> ref_p = 70 70 ;
>> Velocity generation gen_vel =
>> yes gen_temp = 300.0
>> gen_seed = 173529 ; Bonds
>> constraints = all-bonds
>> constraint-algorithm = lincs
>>
>>
>> -- ========================================
>>
>> Justin A. Lemkul
>> Ph.D. Candidate
>> ICTAS Doctoral Scholar
>> MILES-IGERT Trainee
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu <http://vt.edu> | (540) 231-9080
>>
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
>> -- gmx-users mailing list gmx-users at gromacs.org
>> <mailto:gmx-users at gromacs.org>
>>
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>> Please don't post (un)subscribe requests to the list. Use the www
>> interface or send it to gmx-users-request at gromacs.org
>> <mailto:gmx-users-request at gromacs.org>.
>>
>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>>
>>
>>
>> --
>> Moeed Shahamat
>> Graduate Student (Materials Modeling Research Group)
>> McGill University- Department of Chemical Engineering
>> Montreal, Quebec H3A 2B2, Canada
>> Web:
>> http://mmrg.chemeng.mcgill.ca/pages/current-group-members/moeed-shahamat.php
>> Web:http://mmrg.chemeng.mcgill.ca/
>>
>>
> --
> ========================================
>
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>
> ========================================
> --
> gmx-users mailing list gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the www interface
> or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110311/8805e6ef/attachment.html>
More information about the gromacs.org_gmx-users
mailing list