[gmx-users] Re: Umbrella sampling with large pulling distance (larger than half of the box size) (Thomas Schlesier)

[DeChang Li]

>----------------------------------------------------------------------
>
>Message: 1
>Date: Wed, 20 Mar 2013 12:55:45 +0100
>From: Thomas Schlesier <schlesi at uni-mainz.de>
>Subject: [gmx-users] Umbrella sampling with large pulling distance
>        (larger than half of the box size)
>To: <gmx-users at gromacs.org>
>Message-ID: <5149A3C1.2000106 at uni-mainz.de>
>Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
>
>Look for
>pull_geometry = direction_periodic
>This should solve the problem.
>
>Greetings
>Thomas

  Thanks very much for your reply. However, I want to use Umbrella sampling
to get the potential of mean force curve with g_wham. As I know,
pull_geometry = direction cannot do that, so as direction_periodic?


>
>
>Am 20.03.2013 12:00, schrieb gmx-users-request at gromacs.org:
> Dear all,
>
> I want to use Umbrella sampling method to calculate the potential of
> mean force. Unfortunately, the distance of my two groups is very large
> (about 10nm, larger than the half of simulation box size 16nm).
> Without increasing the box size (too large!!), is there any ideas to
> solve this problem?
>
> In another way, is it possible to modify the pulling code (pull.c) to
> force the umbrella sampling method only calculate the distance of the
> groups in the box but not the periodic image?
>
>
>
>------------------------------