Re: eABF - Fullsamples and windows merging

From: yjcoshc_at_gmail.com
Date: Wed Jul 19 2017 - 09:01:40 CDT

Hi,

1. I often set fullsamples to 1000 for 1D reaction coordinate and 500
for 2D. For more information you can read the review:
http://pubs.acs.org/doi/abs/10.1021/jp506633n

2. You can get the colvars source code from
https://github.com/colvars/colvars and compile the abf_integrate in the
colvartools directory, and then run abf_integrate with your grad and
count file to get the final PMF.

Thanks,

yjcoshc

在 2017年07月19日 21:41, philippe Bourly 写道:
>
> Hello !
>
> I am currently learning the eABF method and I have some questions.
>
> First, how to choose the Fullsample value? I read that it is
> recommended to apply the biasing force only after the estimate of the
> free energy has started converging. What is the usual value for this
> parameter?
>
> I also want to split my reaction coordinate in multiple windows.
> Apparently we can merge the windows using the eabf.tcl script. The
> output of this script is a grad and a count file. How to integrate the
> gradient in order to get the PMF?
>
> Thanks,
>
> Philippe
>

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2018 - 23:20:27 CST