On Thu, Jul 20, 2017, Ramin Salimi wrote:
>
> Well, I was planning to calculate the entropy based on the method by Schlitter:
It's been a long time since this came up for me, but the entropy
calculation already done by Amber is more accurate than the Schlitter
approximation. See:
%A I. Andricioaei
%A M. Karplus
%T On the calculation of entropy from covariance matrices of the atomic
fluctuations
%J J. Chem. Phys.
%V 115
%P 6289-6292
%D 2001
It could be a good learning exercise to implement the Schlitter idea and make
comparisons for yourself, but you should first study the existing
quasi-harmonic code so that you understand what it is doing.
Note also that 4ns and only 2000 snapshots are almost certainly well below
what is needed to converge quasiharmonic entropies--but try some experiments
yourself. On the other hand, if the trajectories are not doing what you think
they should, that needs to be investigated. (It seems unlikely that the
strength of the restraint is very important, though.)
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jul 20 2017 - 08:30:03 PDT
