Behavior of explicit water beyond periodic boundaries

From: Dr. Eddie (
Date: Thu Dec 20 2012 - 11:03:02 CST

Hi all,
I using namd to run a solvated protein in a box with PBC. During the runs
my protein tends to leave the box. I understand that it is still solvated,
but it appears it is not explicitly solvated.

When I (using vmd) look at the Hbonds between some side chains' oxygen and
water I get different behavior when the protein is inside the actual box
 vs when it is outside. When it is inside the actual box it will make
explicit Hbonds with the water. Once outside the box it will never do that.

I understand this to mean that while the protein is outside the actual box
(and wrap water is on) the protein interacts with the image o the water but
can't form an actual Hbond with the image.

If I use wrap all then the protein will eventually come back but it will
spend time with some of it outside the box (before enough of it is outside
to wrap) and experience the same lack of Hbond intermittently.

Has anyone else had this sort of issue and come to a resolution? I know I
can use collective variables to keep the protein in the center of the box,
but I worry that may also change the behavior of the protein. Clearly a
much larger box is not realistic, although ideal.

Thanks in advance for the help!!

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:22:52 CST