Just for reference of the Mailing list:
The files were checked with pmemd.cuda in amber16 and amber18, reproducing
the error in amber16 but not in amber18. I think this is then related to
the update done in amber 18 regarding COM restraints and the way the
simulation box is split in GPUs (
http://archive.ambermd.org/201903/0103.html),
but maybe Charlie can comment on that (on CC). Is there any way of porting
this patch to amber16?
Cheers,
El vie., 17 ene. 2020 a las 10:27, Casalini Tommaso (<
tommaso.casalini.chem.ethz.ch>) escribió:
> Dear Stephan,
>
> right now the restraint is aimed at keeping the polymer COM/bilayer COM
> distance at 0 (as shown but the input file that I copied).
>
> I started by simply placing the chain in the simulation box, after
> equilibration the distance was about 4 nm.
>
> The output file with the distances was correctly written, but the
> restraint energy was still ****** and the restraint was ineffective (the
> chain was not moving in the center).
>
> I thought that the distance was too high so I tried to keep the distance
> equal to 3 nm, but the issue remained.
>
> Then I rebuilt the system by placing at the beginning the polymer in the
> middle of the bilayer (distance equal to about 0).
>
> Right now the restraint energy should be reasonable, but it is still
> *******.
>
> The output file with the distances is correctly written, the distance is
> equal to about 0.
>
> If necessary, I can share my files to check if someone can reproduce my
> issue.
>
> ________________________________
> Da: Stephan Schott <schottve.hhu.de>
> Inviato: giovedì 16 gennaio 2020 23:37:43
> A: AMBER Mailing List
> Oggetto: Re: [AMBER] Issues with Umbrella Sampling Simulations involving
> lipid bilayer
>
> So it starts with ****? What is your initial distance then and what values
> do you get in the dump file?
>
> El jue., 16 ene. 2020 a las 13:39, Casalini Tommaso (<
> tommaso.casalini.chem.ethz.ch>) escribió:
>
> > Dear Stephan,
> >
> > thanks a lot for your reply.
> >
> > I am using the pmemd.cuda module of AMBER16, without additional patches.
> >
> > I tried to implement your suggestion and the RESTRAINT energy goes crazy
> > immediately, from step 1 (I put ntpr = 1).
> >
> > As mentioned in the previous mail, what puzzles me is that I used the
> same
> > approach using a different membrane (pure DOPC) and everything went well.
> >
> > Best,
> >
> > Tommaso
> >
> >
> > ________________________________
> > Da: Stephan Schott <schottve.hhu.de>
> > Inviato: giovedì 16 gennaio 2020 13:13:59
> > A: AMBER Mailing List
> > Oggetto: Re: [AMBER] Issues with Umbrella Sampling Simulations involving
> > lipid bilayer
> >
> > Hi Tommaso,
> > Are you using sander, pmemd or pmemd.cuda? What AMBER version are you
> > using, and which patches have been applied? I remember there was an issue
> > with COM restraints, fixed in update 13 of AMBER 18. You could try
> > increasing the rate at which you save the trajectory frames, to take a
> look
> > at the moment when the RESTRAINT energy goes crazy.
> > Cheers,
> >
> > El jue., 16 ene. 2020 a las 10:32, Casalini Tommaso (<
> > tommaso.casalini.chem.ethz.ch>) escribió:
> >
> > > Dear Amber users and developers,
> > >
> > > I am performing Umbrella Sampling simulations to study the permeation
> of
> > a
> > > small polymer fragment inside a POPC/CHL 2:1 lipid bilayer.
> > >
> > >
> > > I built the lipid bilayer with the CHARMM GUI website and I have
> obtained
> > > an equilibrated structure, that I checked with the area per lipid and
> > > bilayer thickness as a function of simulation time.
> > >
> > > I have subsequently put the polymer fragment in the simulation box and
> I
> > > have rebuilt the simulation box with the AddToBox utility, adding water
> > and
> > > ions only along z.
> > >
> > > I ran additional simulations to obtain an equilibrated system
> > > configuration and I started US simulations.
> > >
> > > I use the following input file:
> > >
> > >
> > > Equil
> > > &cntrl
> > > imin = 0,
> > > irest = 1,
> > > ntx = 7,
> > > ntb = 2,
> > > pres0 = 1.0,
> > > ntp = 3,
> > > taup = 2.0,
> > > cut = 10.0,
> > > ntc = 2,
> > > ntf = 2,
> > > temp0 = 310.0,
> > > tempi = 310.0,
> > > ntt = 3,
> > > gamma_ln = 1.0,
> > > nstlim = 20000000,
> > > dt = 0.002,
> > > ntpr = 10000,
> > > ntwx = 10000,
> > > ntwr = 10000,
> > > nmropt = 1,
> > > csurften = 3,
> > > gamma_ten = 0.0,
> > > ninterface = 2,
> > > tol = 0.0000001
> > > /
> > > &wt type = 'DUMPFREQ', istep1 = 100 /
> > > &wt type = 'END', /
> > > DISANG = ref_COM.rst
> > > DUMPAVE = Pull_dist.dat
> > > LISTIN = POUT
> > > LISTOUT = POUT
> > > /
> > > /
> > > &ewald
> > > skinnb = 3.0,
> > > /
> > >
> > > This is the ref_COM.rst file:
> > >
> > >
> > > &rst
> > > iat=-1,-1,
> > > r1=-99.0,
> > > r2=0.0,
> > > r3=0.0,
> > > r4=99.0,
> > > rk2=2.5,
> > > rk3=2.5,
> > > iresid=0,
> > > fxyz=0,0,1,
> > > outxyz=1,
> > >
> > >
> >
> igr1=1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,
> > >
> > >
> >
> igr2=130,264,398,606,888,1022,1156,1364,1498,1632,1766,1900,2034,2168,2376,2510,2866,3000,3208,3342,3476,3610,3744,3878,4086,4294,4576,4710,4844,4978,5112,5320,5602,5736,6018,6226,6360,6568,6850,6984,7192,7326,7534,7668,7802,7936,8144,82
> > >
> > >
> >
> 78,8412,8546,8754,8888,9022,9230,9512,9646,9780,9914,10048,10182,10316,10524,10954,11088,11222,11356,11490,11624,11758,11966,12248,12382,12516,12650,12858,12992,13126,13260,13468,13676,13884,14018,14226,14434,14568,14702,14836,14970,1510
> > >
> > >
> >
> 4,15238,15520,15654,15862,15996,16204,16338,16472,16606,16740,17022,17156,17438,17572,17780,17914,18270,18552,18686,18820,18954,19088,19296,19430,19786,19920,20054,20188,20322,20456,20590,20724,20858,20992,21200,
> > > /
> > >
> > > where atoms 1-63 belong to my polymer fragment, while atoms of igr2 are
> > > the N31 atoms of the polar heads (I essentially follow the approach
> > > explained by Dr. Dickson Callum in his tutorial). I checked the numbers
> > and
> > > they are correct.
> > >
> > > For some unknown reasons, I obtained this:
> > >
> > >
> > > NSTEP = 10000 TIME(PS) = 10540.000 TEMP(K) = 309.33 PRESS =
> > > -116.3
> > > Etot = ************** EKtot = 34089.0039 EPtot =
> > > **************
> > > BOND = 2925.3688 ANGLE = 12094.6034 DIHED =
> > > 7501.6413
> > > 1-4 NB = 2919.0577 1-4 EEL = -42192.4792 VDWAALS =
> > > 5238.4991
> > > EELEC = -110714.2666 EHBOND = 0.0000 RESTRAINT =
> > > **************
> > > EAMBER (non-restraint) = -122227.5755
> > > EKCMT = 9433.8540 VIRIAL = 10678.6782 VOLUME =
> > > 495566.9126
> > > SURFTEN =
> > > -52.4235
> > > Density =
> > > 1.0067
> > >
> > >
> >
> ------------------------------------------------------------------------------
> > >
> > > NMR restraints: Bond =********* Angle = 0.000 Torsion =
> > 0.000
> > >
> > >
> >
> ===============================================================================
> > >
> > > NSTEP = 20000 TIME(PS) = 10560.000 TEMP(K) = 311.67 PRESS =
> > > 187.4
> > > Etot = ************** EKtot = 34346.3750 EPtot =
> > > **************
> > > BOND = 2985.2622 ANGLE = 12157.1242 DIHED =
> > > 7567.3432
> > > 1-4 NB = 2901.6475 1-4 EEL = -42177.9928 VDWAALS =
> > > 5504.9903
> > > EELEC = -110692.8191 EHBOND = 0.0000 RESTRAINT =
> > > **************
> > > EAMBER (non-restraint) = -121754.4445
> > > EKCMT = 9541.5155 VIRIAL = 7537.2160 VOLUME =
> > > 495324.5851
> > > SURFTEN =
> > > 86.8556
> > > Density =
> > > 1.0072
> > >
> > >
> >
> ------------------------------------------------------------------------------
> > >
> > > NMR restraints: Bond =********* Angle = 0.000 Torsion =
> > 0.000
> > >
> > >
> >
> ===============================================================================
> > >
> > > NSTEP = 30000 TIME(PS) = 10580.000 TEMP(K) = 307.85 PRESS =
> > > -10.4
> > > Etot = ************** EKtot = 33925.5469 EPtot =
> > > **************
> > > BOND = 3003.1912 ANGLE = 12038.2158 DIHED =
> > > 7478.7713
> > > 1-4 NB = 2874.2811 1-4 EEL = -42133.3378 VDWAALS =
> > > 5436.8241
> > > EELEC = -110538.2867 EHBOND = 0.0000 RESTRAINT =
> > > **************
> > > EAMBER (non-restraint) = -121840.3409
> > > EKCMT = 9409.4898 VIRIAL = 9520.6799 VOLUME =
> > > 495200.1274
> > > SURFTEN =
> > > -193.4285
> > > Density =
> > > 1.0075
> > >
> > >
> >
> ------------------------------------------------------------------------------
> > >
> > > NMR restraints: Bond =********* Angle = 0.000 Torsion =
> > 0.000
> > >
> > > If I do not add a restraint, the system behaves nicely.
> > >
> > > What puzzles me is that I used the same protocol and approach for the
> > same
> > > polymer fragment in a pure DOPC membrane and I had no problems, and the
> > > obtained results are consistent.
> > >
> > > Can you provide some suggestions?
> > >
> > > I thank you in advance for your support!
> > >
> > > Tommaso
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> > --
> > Stephan Schott Verdugo
> > Biochemist
> >
> > Heinrich-Heine-Universitaet Duesseldorf
> > Institut fuer Pharm. und Med. Chemie
> > Universitaetsstr. 1
> > 40225 Duesseldorf
> > Germany
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
>
> --
> Stephan Schott Verdugo
> Biochemist
>
> Heinrich-Heine-Universitaet Duesseldorf
> Institut fuer Pharm. und Med. Chemie
> Universitaetsstr. 1
> 40225 Duesseldorf
> Germany
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Stephan Schott Verdugo
Biochemist
Heinrich-Heine-Universitaet Duesseldorf
Institut fuer Pharm. und Med. Chemie
Universitaetsstr. 1
40225 Duesseldorf
Germany
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jan 20 2020 - 07:30:02 PST
