James,
Ninterface should be set to the number of interfaces in your system. In your
case, it should be ninterface = 4. Basically, surface tension is implemented
into the pressure coupling in such a way that the reference surface tension
value, 'gamma_ten', is divided by the number of interfaces in the
calculation of the reference pressure for pressure coupling.
For some initial reading on constant surface tension, I suggest Zhang, Y. et
al. J. Phys. Chem. 1995 and Feller, et al. J. Phys. Chem. 1995.
I don't know what reference surface tension value to use for your system. I
think it will be larger than a single bilayer, but it won't necessarily be
twice a single bilayer system's surface tension.
I know this type of system has been done before using other force fields.
However, our group has not tested multiple interfaces in a system like you
describe. I am also interested in how this system behaves.
Ben Madej
Walker Molecular Dynamics Lab
-----Original Message-----
From: Ross Walker [mailto:ross.rosswalker.co.uk]
Sent: Monday, July 16, 2012 10:59 AM
To: 'AMBER Mailing List'
Subject: Re: [AMBER] AMBER12 - ninterface for double bilayer system
Hi James,
Ben Madej might be able to comment more here since he was the one who
implemented the constant surface tension code. My understanding is that
setting the number of interfaces to 4 should be sufficient - it is basically
a scale factor for the actual surface tension value.
That said I don't think we have ever actually tried multiple bilayers so
this will definitely be an experiment. I believe the surface tension should
probably just be the same as a single bilayer but of course that will depend
if the distance between the bilayers is small such that there is substantial
interaction between the two. Either way I'd be interested to hear what you
find when you actually try it.
All the best
Ross
> -----Original Message-----
> From: James Murdock [mailto:james.murdock.86.gmx.com]
> Sent: Monday, July 16, 2012 3:31 AM
> To: amber.ambermd.org
> Subject: [AMBER] AMBER12 - ninterface for double bilayer system
>
> Dear Amber users,
>
> I'm currently interested in the simulation of a system composed by
> two bilayers separated by a water slab between those two bilayers and
> with water slabs on the simulation cell boundaries. As I intend to use
> LIPID11, a surface tension dependent FF for this system, I believe
> that the ninterface value should be adjusted accordingly to the number
> of water/lipid interfaces, following the statement on page 26 of the
> AMBER12 manual:
>
> ninterface
> Number of interfaces in the periodic box. There must be at least two
> interfaces in the periodic box. Two interfaces is appropriate for a
> lipid bilayer system and is the default value.
> So, if for a single bilayer system ninterface = 2, then for a double
> bilayer system ninterface = 4. What are your thoughts concerning this
> change?
> I am yet to discover if the surface tension value should be changed
> from the value appropriate for a single bilayer system.
>
> Regards,
> JM
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 16 2012 - 16:30:02 PDT
