Re: [AMBER] Issues with Umbrella Sampling Simulations involving lipid bilayer

From: Casalini Tommaso <tommaso.casalini.chem.ethz.ch>
Date: Thu, 16 Jan 2020 12:37:26 +0000

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: giovedý 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>) escribiˇ:

> 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
Received on Thu Jan 16 2020 - 05:00:01 PST
Custom Search