AMBER Archive (2009)

Subject: Re: [AMBER] surface binding protein

From: Jason Swails (
Date: Sat Oct 31 2009 - 07:35:44 CDT


We'd have to see your input files to provide more definitive help, but I may
be able to explain what's happening. If you're simulating at either
constant volume or constant pressure (ntb=1 or ntb=2, respectively), with
iwrap=1, then what you may be seeing is an artifact of the ligand being
'wrapped' back to the other side of the box. iwrap=1 maps molecules that
leave one side of the periodic box to the other so that during the course of
the simulation, the visualized periodic box doesn't change. If this did not
occur, it would appear as though the waters exploded from the protein, but
this is actually due to the diffusion of water molecules into adjacent
boxes. However, the way simulations are handled, every periodic box is
included in the calculation, so even though it may appear as though the
ligand jumps to the other side of the box, the simulation still acts as
though there is a ligand bound to the substrate (and there is actually a
substrate bound to the apparently free ligand as well, since each of those
parts now belong to 'different' complexes in different periodic boxes).

You can tell that this is the effect if what you see is the ligand "jumping"
many angstroms away from the receptor. This is completely normal and does
not affect your simulation at all. You can use ptraj to recenter and image
the protein inside the box.

Good luck!

On Sat, Oct 31, 2009 at 1:28 AM, sadaf iqbal <> wrote:

> Hello all,
> I am doing simulation of a protein in which substrate is bound to the
> surface of protein. During MD production run, I have encountered a problem
> that substrate move away far and far from the binding surface as the
> simulation proceed. Can anybody help me to tackle this problem associated
> with the simulation of surface binding substrate/protein complex. I have
> done simulation with substrate/protein complex in which substrate is in
> pocket/subsites and never encountered this problem.
> Any help in this regard will be highly appreciated.
> Sadaf Iqbal
> Computational Medicinal Chemistry Unit
> Dr. Panjwani Center for Molecular Medicine and Drug Research
> New Email names for you!
> Get the Email name you've always wanted on the new @ymail and @rocketmail.
> Hurry before someone else does!
> _______________________________________________
> AMBER mailing list

Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
AMBER mailing list