From: Hannes Loeffler (Hannes.Loeffler_at_stfc.ac.uk)
Date: Thu Nov 04 2010 - 08:42:59 CDT
On Thu, 4 Nov 2010 06:49:40 -0500
Chris Harrison <charris5_at_gmail.com> wrote:
> In the case of a protein system, which contains heterogeneous pairs
> that are not a regular lattice, <> over all h-bond pairs does not
> seem particularly meaningful. In such an example, you should most
> likely restrict <> to a particular h-bonded pair over your
> trajectory, unless you are examining for example the water-to-water
> h-bonding properties of water solvating the protein.
Certainly the way to go but the result may not be too
statistically meaningful for a single pair. It may be useful to work
with multiple time origins (sliding window). Of course, after checking
that they are not (too strongly) correlated. It may also be necessary
to check whether the environment of a particular HB pair does not change
the properties of the nature of the HB in the course of a simulation.
Cheers,
Hannes.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:56:19 CST