[gmx-users] protein folding / pbc

[Mark Abraham]

On 11/04/2012 1:24 AM, Shi, Huilin wrote:
> Hi All,
>
> I have a question about periodic boundary conditions (pbc) when 
> running a simulation to unfold a protein from its native structure.
>
> I set up pbc with the starting structure which is compact.
> When defining the box we use editconf and set up the -d 1.0 as the 
> distance from the protein to the box edge.
> It is said that "a protein should never see its periodic image".
> If during the simulation the protein starts unfolding, the minimum 
> distance from the protein to the box edge will no longer be 1.0 nm.
> And it is possible that the length of the protein is even longer than 
> the box dimension and the protein may "see" its periodic image in the 
> neighbored box.
> Is this going to be a problem?

Yes, if you don't allow for it from the start. Some test simulations 
with implicit solvent are probably priceless here.

> I am also confused that it is also said that "(parts of) the 
> molecule(s) diffuse out of the box is not a problem".
> Is this conflicted with "a protein should never see its periodic image"?

Sometimes - can depend on orientation of molecule wrt box.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120411/2b5c4061/attachment.html>