On Thu, 13 Apr 2017 09:27:20 +0000
吴萌 <wumeng.shanghaitech.edu.cn> wrote:
> Dear All,
>
> I am trying to use the AMBER umbrella sampling method to get the
> free energy of transfer profile for a substrate molecule through a
> channel protein. And I have referred
> https://github.com/callumjd/AMBER-Umbrella_COM_restraint_tutorial ,
> but several problems have trapped me due to my system is different.
>
> 1) If I put one substrate molecule close to the protein channel
> entrance, how can I define its location, I mean such as Z=0(need it
> to get r2=? in the pull.RST)
It looks like you are defining a distance between the groups igr1 and
igr2 so I would think that you set r2 to the current distance.
> 2) Also the distance restraint file, I
> use igr1=substrate atom number, how should I difine igr2=? use
> residue near the protein channel exit? If yes, use one or several?
The tutorial you refer to uses the center-of-mass (COM) of all head
groups which probably, on average, means in the center of the box with
respect to the x and y distance (or bilayer actually if it is the head
groups of both leaves). The tutorial only considers the distance between
the bulk and the center of the bilayer as the order parameter (and so
assumes symmetry).
I suspect the COM of each group will only be computed once at the
beginning of the simulation. What I would do is to introduce a
virtual/dummy atom at a convenient location (maybe use ibelly to fix
it in space) and use that as reference point.
Plumed would allow you to conveniently do all this but this is only
available for sander. There is some similar functionality (US, SMD) in
the NFE code (previously known as NCSU) but I can't see in the manual
that you could define virtual sites or consider only one specific
component of the cartesian coordinate.
Cheers,
Hannes.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Apr 13 2017 - 06:30:02 PDT
