Hi,
On Thu, Feb 27, 2014 at 8:11 AM, Yew Mun Yip <yipy0005.gmail.com> wrote:
>
> I have read previous posts about setting iwrap=1 for running long
> simulations with pmemd. But the manual indicates that running long
> simulations with that setting will also result in consequences, therefore
> with all the conflicting viewpoints, should I or should I not set iwrap=1?
> What are the consequences if iwrap=1 is set for long simulations? If I
> don't set iwrap=1, what can I do to continue the simulation correctly?
>
As mentioned in the manual, coordinate "wrapping" is for visual and
numerical convenience in simulations with periodic boundary conditions. The
only consequence is that molecules that drift out of the main unit cell and
become wrapped take discontinuous paths. This can result in molecular
complexes (such as a DNA duplex) appearing "split" in the resulting
trajectory, although this is just a visual artifact and does not affect the
forces/energies calculated during the simulation. It also means you
probably cannot use the trajectory as-is to calculate things like the
self-diffusion coefficient using e.g. the 'diffusion' command in cpptraj.
However, cpptraj has many commands that allow you to manipulate imaged
trajectories to restore "split" complexes (autoimage), re-center on a
molecule of interest (autoimage, image), or obtain the trajectory without
wrapping (unwrap), so in general it is best to always use iwrap=1 in
simulations with PBC and post-process with cpptraj as needed. In addition,
I recommend using NetCDF trajectories and restarts (ioutfm=1 and ntxo=2
respectively) since they have higher precision and take longer to overflow
than the corresponding ASCII formats (and NetCDF trajectories are much
faster to process).
Hope this helps,
-Dan
>
> Many thanks, because I really hope to clear this once and for all. =D
>
> --
> Regards
> Yip Yew Mun
> Graduate, PhD
> Chemistry and Biological Chemistry
> School of Physical and Mathematical Sciences
> Nanyang Technological University
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Feb 27 2014 - 08:00:03 PST
