Re: ABF question

From: Jérôme Hénin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Fri Dec 03 2010 - 10:45:24 CST

Hi Parisa,

In principle it is true, but diffusion can be slow, and diffusive
sampling can be inefficient. A way to speed things up is to split the
channel into windows, e.g. 4 windows each 10-Angstrom long. You can
then start 4 separate simulations with adequate initial positions of
the ion, and the full channel will be sampled faster.

Jerome

On 3 December 2010 16:16, Parisa Akhski <Parisa.Akhshi_at_chem.queensu.ca> wrote:
> Dear NAMD users
>
> I need to get the PMF profile of a channel ion along the channel with RC=-20
> to +20.In using ABF (adaptive biasing force) method, we believe that:
>
> "Eventually, as the underlying free energy surface is canceled by the
> adaptive bias, evolution of the system along the reaction coordinate is
> governed mainly by diffusion."
>
> To me that means it doesn't matter where I put the first position of my
> channel ion at the beginning as long as it is between -20 and +20. I mean
> even if I don't put it in the middle, it should have the chance to go
> through all the reaction pathway finally with almost the same number of
> sampling everywhere.
>
> The reason I am asking this is that the ion has been spending much time in
> half of the RC and never has been in the other half so far. Just want to
> confirm that the begging position of the ion really doesn't matter as long
> as it is in the RC. Is that true?
> Thanks,
> Parisa

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:23:27 CST