Hi Jason,
Thanks for your response. So something did go wrong and I got the “Coordinate resetting (SHAKE) cannot be accomplished,
>
> deviation is too large” error with iwrap and harmonic restraints together.
So I turned off iwrap keeping everything else the same, assuming a restrained molecule will not be diffusing much without any problems in the restraint energy. However, at the end of each restart, the molecule is translated (I can see in VMD that the coordinates at the end the first restart and at the start of the second one are not the same). And when I restart after 250 psec, the restraint energy increases significantly from the last step of the previous restart (1.0082) to the first step of the current restart (112371.5346). At the beginning of the 3rd restart, every more strangely, the substrate is separated from the enzyme (I don’t think it’s simply diffusion because it’s still bound tightly at the end of the 2nd restart). The restraint is not only on the substrate by the way, both substrate and the a few residues of the binding site is restrained. On the other hand, when I restart at every 1ns, during the first 1 ns everything’s fine (coordinates and energies) but the second rst won’t start.
I thought because I set ig=-1 (for Langevin thermostat), could I actually be assigning new velocities at each restart? But it shouldn’t be the case because ntx=5 and irest=1 and it should be reading the velocities from the previous restart. Could it be the way output is written at the end of each rst or the way input is read? Any ideas on what I'm missing?
Thanks! I appreciate your help!
Aysegul
My mdin file:
FOR HARMONICALLY RESTRAINED MD Simulation (@300K, NPT ensemble, 1 ns, no restraint)
&cntrl
! iwrap=1, !the crd written to the rst and crd files will be wrapped into a primary box.
irest=1, !restart calculation
ntx=5, !X and V are read formatted, box info read if ntb>0. V used only if irest=1
temp0=300.0, !Temperature
ntt=3, gamma_ln=1.0,ig=-1, !Langevin dynamics for temperature regulation
ntpr=5, !mdout/mdinfo written in every 5 steps
ntwr=5, !rst written in every 5 steps
ntwx=5, !mdcrd written in every 5 steps
ntb=2, !constant pressure
ntp=1, !constant pressure dynamics, md with isotropic position scaling
ntf=2, !force evaluation (bond interactions involving hydrogens omitted)
ntc=2, !SHAKE (bonds involving hydrogens are omitted)
ntr=1, restraintmask=':1-7,78,144,153,156-160,175-181,186 & @CA', restraint_wt=1.0 !kcal/mol/A2
nstlim=125000, dt=0.002,
&end
END
________________________________________
From: Jason Swails [jason.swails.gmail.com]
Sent: Thursday, August 18, 2011 8:50 PM
To: AMBER Mailing List
Subject: Re: [AMBER] Using iwrap with harmonic restraints
Only if the restrained molecule were to move out of your primary box (it
would then get reimaged). I don't think this would be an issue until you
re-started the simulation (if you kept the same reference structure, of
course).
However, this is unlikely to happen to restrained molecules, so I'm guessing
you'll be fine here. The upside is if anything *does* go wrong, it will go
wrong in a spectacularly noticeable way.
HTH,
Jason
On Thu, Aug 18, 2011 at 8:44 PM, Ozen, Aysegul <Aysegul.Ozen.umassmed.edu>wrote:
> Dear Amber users,
>
> I've been using iwrap=1 in my long simulations as recommended in the Amber
> 10 manual without any issues so far. However, I need to set up a new
> simulation where a subset of residues has to be restrained harmonically in
> the production with respect to the minimized coordinates. I'm restarting at
> every 1 ns and at each restart the reference for the restraint is the same
> minimized structure. My question is would wrapping the coordinates into the
> primary box interfere with the way restraining works? Here's my mdin:
>
>
> &cntrl
>
> iwrap=1, !crd and rst wrapped into a primary box
> irest=1, !restart calculation
> ntx=5, !X and V are read formatted, box info read as
> ntb=2
>
> !constant T
> temp0=300.0,
> ntt=3, gamma_ln=1.0, ig=-1,
>
> !constant P
> ntb=2,
> ntp=1,
>
> !SHAKE
> ntc=2,
> ntf=2,
>
> !restraint
> ntr=1, restraintmask=':1-7,78,144,153,156-160,175-181,186 & @CA',
> restraint_wt=1.0 !kcal/mol/A2
>
> !outputs
> ntpr=500,
> ntwr=50000,
> ntwx=500,
>
>
> nstlim=500000, dt=0.002
>
> &end
> END
>
>
> Thanks!
> Aysegul
>
> _______________________________________________
> 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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 22 2011 - 09:00:03 PDT
