AMBER Archive (2004)

Subject: RE: AMBER: MD simulation of a DNA decamer

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Tue Mar 09 2004 - 12:07:35 CST

Dear Thomas,

The original DNA tutorial on the AMBER website was considerably dated and
not successfully updated for newer versions of AMBER. There is now a new,
more comprehensive DNA tutorial in place of the old one. Check
http://amber.scripps.edu/tutorial/.

I'm not sure what is happening with your temperature but you could try doing
constant VOLUME heating (tempi=0, temp0=300) followed by constant pressure
equilibration (tempi=300, temp0=300) this should give you a stable
temperature. The tutorial now uses Langevin dynamics to control the
temperature (new method implemented in AMBER8) but if you just use the weak
coupling algorithm (NTT=1) with the default settings you should get a stable
temperature. Let me know how you get on.

All the best
Ross

/\
\/
|\oss Walker

| Department of Molecular Biology TPC15 |
| The Scripps Research Institute |
| Tel:- +1 858 784 8889 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk/ | PGP Key available on request |

> -----Original Message-----
> From: owner-amber_at_scripps.edu
> [mailto:owner-amber_at_scripps.edu] On Behalf Of Thomas Steinbrecher
> Sent: 09 March 2004 02:10
> To: amber_at_scripps.edu
> Subject: AMBER: MD simulation of a DNA decamer
>
> Dear AMBER users,
>
> I am trying to simulate a DNA decamer in explicit solvent
> using periodic boundary conditions with neutralizing Na+
> ions, similar to the simulations described in Srinivasan et
> al. ACS 1998.
>
> I loosely followed the equilibration procedure detailed in
> the first tutorial on the AMBER homepage (details see
> below), but I run into trouble:
>
> After equilibration, my production runs yielded system
> temperatures of ~360K. I used tautp=1.0 and temp0=300 as
> parameters. Changing tautp to 0.5 or temp0 to 270 let the
> temperature level off at ~330K. Can someone tell me if this
> behaviour is to be expected? I was wondering if something
> in my equilibration procedure is wrong or if this is a
> known problem in DNA simulations. Is further lowering the
> target temperature temp0 a valid workaround if it results
> in a system temperature of ~300K or will this cause a more
> unnatural trajectory?
>
> Many thanks in advance for any help and comments,
>
> kind regards,
>
> Thomas
>
>
> Details of my equilibration procedure:
>
> I do an initial minimization, followed by 25ps of constant
> pressure equilibration with restrained solute (yields a
> density of ~1) and subsequent minimization without
> restraints. I then equilibrate my system for 50-400 ps at
> constant volume conditions to equilibrate the system
> temperature (longer runs don't seem to change the resulting
> temperature in the production runs), followed by 25 ps of
> constant pressure simulation to equilibrate density. I use
> time constants of 0.2 ps for tautp and taup during
> equilibration. I use cut=12.0, because cut=9.0 leaded to
> EWALD bomb failures.
>
> My mdin file for production runs
>
> &cntrl
> imin = 0, nmropt = 0, ntx = 5, irest = 1,
> ntrx = 1, ntxo = 1, ntpr = 5000, ntave = 0,
> ntwr = 5000, iwrap = 0,
>
> ntwx = 5000, ntwv = 0, ntwe = 0, ioutfm = 0,
> ntwprt = 0, idecomp = 0,
>
> ntf = 2, ntb = 2, dielc = 1, cut = 12.0,
> scnb = 2.0, scee = 1.2, nsnb = 10, ipol = 0,
>
> ibelly = 0, ntr = 0,
>
> nstlim = 500000, nscm = 1000, t = 0.0, dt = 0.002,
>
> temp0 = 300.0, tempi = 100, ig = 72352,
> heat = 0.0, ntt = 1, vrand = 0, tautp =
> 1.0,
> vlimit = 20.0,
>
> ntc = 2, tol = 0.00001,
>
> ntp = 1, pres0 = 1.0, comp = 44.6, taup = 1.0,
> npscal = 1,
> &end
>
> If anybody is interested, I will of course send my other
> input and output files.
>
> Thomas Steinbrecher
> Institut für physikalische Chemie
> Albertstr. 23a
> 79108 Freiburg
> --------------------------------------------------------------
> ---------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu