On Sun, Feb 7, 2016 at 6:12 AM, Weinzierl, Robert O J <
r.weinzierl.imperial.ac.uk> wrote:
> Dear Dan,
>
> Thank you for your (speculative) reply. Sometimes it is good to know that
> the answer is either unclear or unknown.
>
> The only thing I was unclear about in your explanation is the effect of
> this selective boost (" ... if the average energy of your "folded"
> conformation lies below the boost energy threshold and the average energy
> of your "unfolded" conformation lies above it ...").
> Would this, in practical terms, imply that the folded structure is likely
> to be overrepresented in the trajectory, or do I have this concept the
> wrong way round?
>
I would say you have it the wrong way around. Accelerated MD will
increase the representation of high-energy structures in your trajectory
(pretty substantially depending on your boost parameters). This is why
reweighted FE surfaces from aMD simulations tend to be noisy -- you didn't
sample enough low-energy structures before wandering off to high-energy
parts of phase space.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Feb 07 2016 - 13:00:04 PST
