AMBER Archive (2006)

Subject: RE: AMBER: Restart problem on PMEMD with position restraint.

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Mon Sep 18 2006 - 16:57:58 CDT

Dear Chunhu,

The approach you outline below should be perfectly acceptable and is a way
to get around what is really a limitation of amber's restraint treatment for
the NPT ensemble. It is more costly in terms of cpu time and human setup
time which if they are not an issue is fine but if this is an issue then you
may want to consider restraining to the input coordinates rather than the
minimised coordinates. It really depends what you are using the restraints
for. Do you strictly want to restrain to the minimised coordinates or are
you just using the restraints to stop the solute changing too much during
the initial equilibration phase.

All the best
Ross

/\
\/
|\oss Walker

| HPC Consultant and Staff Scientist |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

> -----Original Message-----
> From: owner-amber_at_scripps.edu
> [mailto:owner-amber_at_scripps.edu] On Behalf Of Chunhu Tan
> Sent: Monday, September 18, 2006 14:11
> To: amber_at_scripps.edu
> Subject: RE: AMBER: Restart problem on PMEMD with position restraint.
>
> Thank you all for reply!
> Now I think the problem is that I used the minimized structure as the
> reference in my both NPT runs. In fact, I encountered this problem
> before, and I dealt with it in this way: in the second NPT run, I use
> the same reference, but with smaller restraint factor, and if needed,
> shorter time step, 0.001. Then, after 200ps or more, the structure was
> as if "pulled" back to the reference. Usually in the third NPT run, I
> can use dt = 0.002 and a larger factor without Vlimit error. But I am
> not sure this is correct. What's your opinion?
>
> Chunhu
>
> -----邮件原件-----
> 发件人: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] 代表
> David A. Case
> 发送时间: Monday, September 18, 2006 12:10 PM
> 收件人: amber_at_scripps.edu
> 主题: Re: AMBER: Restart problem on PMEMD with position restraint.
>
> On Mon, Sep 18, 2006, Chunhu Tan wrote:
> >
> > I got a Vlimit error when I wanted to restart a PMEMD simulation
> with
> > position restraint. My system is protein G solvated in a TIP3P water
> box
> > with 11 angstroms buffer. After the minimization, I first
> run 20ps NVT
> > simulation, then 20ps NPT simulation. All these were OK, but when I
> > wanted to restart the NPT simulation, I got the Vlimit error, even I
> > used exact the same input file as the first NPT run.
>
> What are you using as REFC for the run that failed? This must be the
> restrt
> file from the previous NPT run, *not* just the crystal structure (or
> whatever)
> coordinates. The reason is that the first NPT run scales the
> coordinates,
> but not the reference coords. So, when you start the second NPT run,
> you
> need to reference to scaled coords.
>
> This is a (minor?) limitation of the Amber equilibration scheme. You
> might
> get around it by reducing the strength of your restraints by
> a factor of
> 10
> or 50 (i.e. to use a restraint weight of 1-5 kcal/mol-A**2).
> A value of
> 50 is way too high in any event.
>
> ...hope this helps...dac
>
> --------------------------------------------------------------
> ---------
> 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
>

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