Hello,
This is a little deceiving. The actual bug resides in Amber 10, NOT in
Amber 11. The issue is that during dynamics, the forces are calculated, the
energies are printed, the dynamics is propagated, then the exchange attempt
is made so that the positions of the particles do not *quite* match up with
the energy reported in the mdout file (in fact, they're a full time step
behind). This was noticed well after the Amber 10 release, so the Amber 11
code was modified to calculate the energies and forces again right before
the exchange attempt was made.
Thus, the potential energies used in the exchange attempt will *not* match
the potential energies printed out to the mdout on that step. They will,
however, match the potential energies for the step directly afterwards (you
can verify this by setting ntpr = 1 in the input file), which is the "true"
potential energies for the structures attempting to swap. In any case, this
"bug" was considered to be mild enough not to warrant a large bugfix in
Amber 10 (since not only would the code need to be changed, so would all of
the tests as well).
Hope this helps,
Jason
On Tue, Oct 25, 2011 at 6:24 AM, Mariarosaria Ferraro <
mariarosaria.ferrar2.studio.unibo.it> wrote:
>
>
>
> ________________________________
> Hi All,
> like Amber tutorial A7, I tried to carry out REMD simulation in Amber 11
> (as it's specified, tutorial should also work with this version). In ''
> mdout '' files I noticed that '' EPtot '' values for the last step of MD
> before swapping were different from '' myEptot '' ones used to calculate the
> exchange probability and it looks inconsistent.
> So I made at once Amber tests to check software compilation for accuracy
> and they have been passed; furthermore the same tutorial was tried in Amber
> 10 and no mismatches between these values were found.
> Can I use Amber 11 to simulate my system anyway or do you think Amber 10
> would work better?
> Fragments of the 2 '' mdout '' files for the first replica are wrote out
> below as you can check '' EPtot '' and '' myEptot '' values on your own:
>
> Amber 10 results:
>
>
> ===============================================================================
>
> NSTEP = 500 TIME(PS) = 203.000 TEMP(K) = 293.90 PRESS =
> 0.0
> Etot = 83.0526 EKtot = 79.4283 EPtot =
> 3.6242
> BOND = 22.4261 ANGLE = 38.9603 DIHED =
> 86.4959
> 1-4 NB = 22.7479 1-4 EEL = 632.8255 VDWAALS =
> -25.9882
> EELEC = -707.9128 EGB = -66.2172 RESTRAINT =
> 0.2868
> EAMBER (non-restraint) = 3.3375
> TEMP0 = 269.5000 REPNUM = 1 EXCHANGE# =
> 3
>
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.287
>
> ===============================================================================
> REMD: myEptot= 3.6242 myTargetTemp= 269.50 mytemp= 293.90
> ==========================REMD EXCHANGE
> CALCULATION==========================
> Exch= 4 RREMD= 0
> Replica Temp= 269.50 Indx= 1 Rep#= 1 EPot= 3.62
> Partner Temp= 300.00 Indx= 2 Rep#= 2 EPot= 21.91
> Not controlling exchange.
> Rand= 0.456420E+00 MyScaling= -1.00 Success= F
> ========================END REMD EXCHANGE
> CALCULATION========================
>
> Amber 11 results:
>
>
> ===============================================================================
>
> NSTEP = 500 TIME(PS) = 203.000 TEMP(K) = 238.75 PRESS =
> .0
> Etot = 72.6662 EKtot = 64.5229 EPtot =
> 8.1433
> BOND = 18.1102 ANGLE = 47.0878 DIHED =
> 83.7739
> 1-4 NB = 23.3925 1-4 EEL = 638.6269 VDWAALS =
> -25.4731
> EELEC = -704.7544 EGB = -72.6212 RESTRAINT =
> .0008
> EAMBER (non-restraint) = 8.1425
> TEMP0 = 269.5000 REPNUM = 1 EXCHANGE# =
> 3
>
> ------------------------------------------------------------------------------
>
> NMR restraints: Bond = .000 Angle = .000 Torsion = .001
>
> ===============================================================================
> REMD: myEptot= 8.5716 myTargetTemp= 269.50 mytemp= 238.75
> ==========================REMD EXCHANGE
> CALCULATION==========================
> Exch= 4 RREMD= 0
> Replica Temp= 269.50 Indx= 1 Rep#= 1 EPot= 8.57
> Partner Temp= 300.00 Indx= 2 Rep#= 2 EPot= 20.02
> Not controlling exchange.
> Rand= .336937E-01 MyScaling= -1.00 Success= F
> ========================END REMD EXCHANGE
> CALCULATION========================
>
>
> Thank you in advance.
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Oct 25 2011 - 06:30:03 PDT
