AMBER Archive (2008)

Subject: Re: AMBER: trouble interpreting <P2> time correlation function data

From: David A. Case (case_at_scripps.edu)
Date: Sun Jul 06 2008 - 10:04:30 CDT

On Sun, Jul 06, 2008, Sally Pias wrote:
>
> Specifically, I find that every residue shows "volatility" in <P2>
> toward the end of the simulation, regardless of simulation length and
> regardless of the residue's relative mobility.

This is exactly what you should expect. One rule of thumb is that you can
only reliably calculate time-correlation functions up to about 1/10 of the
length of the underlying trajectory. For real precision, even longer
trajectories are needed.

Here are a couple of papers that discuss this:

%A R. Zwanzig
%A N.K. Ailawadi
%T Statistical error due to finite time averaging in computer experiments
%J Phys. Rev.
%V 182
%P 280-283
%D 1969

%A V. Wong
%A D.A. Case
%T Evaluating Rotational Diffusion from Protein MD Simulations
%J J. Phys. Chem. B
%V 112
%P 6013-6024
%D 2008
 (check references in this paper)

...hope this helps...dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu