- Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]

From: David Cerutti <dcerutti.mccammon.ucsd.edu>

Date: Tue, 21 Aug 2007 17:07:32 -0700 (PDT)

Update: it seems I have found the solution ot my problem. If anyone is

interested, the Octahedral box in AMBER is rotated away from the

orientation I was hoping for, which would be to have square faces on the

XY/XZ/YZ planes. To get it back into this orientation, one must apply the

transformation matrix:

cos1 = cos(PI/4.0);

sin1 = sin(PI/4.0);

cos2 = sqrt(2.0)/sqrt(3.0);

sin2 = 1.0/sqrt(3.0);

R = [

cos1*cos2 -cos1*sin2 sin1

-sin1*cos2 sin2*sin1 cos1

-sin2 -cos2 0.0

];

to the coordinates in the AMBER coordinates file. (This is the reverse

rotation as can be found in the ptraj source code actions.c.)

Then, the dimensions given in the AMBER coordinates file apply to the

distance between two hexagonal faces. Multiply those by 2.0/sqrt(3.0) to

obtain the distance between two square faces. Then, all you need to do to

tile the box is translate the unit cell to different lattice positions:

1 0 0

0 1 0

0 0 1

-1 0 0

0 -1 0

0 0 -1

1/2 1/2 1/2

1/2 -1/2 1/2

-1/2 1/2 1/2

-1/2 -1/2 1/2

1/2 1/2 -1/2

1/2 -1/2 -1/2

-1/2 1/2 -1/2

-1/2 -1/2 -1/2

Note that replicating the octahedron at one of the (1/2 1/2 1/2) lattice

positions, then re-imaging all particles as if there were an orthonormal

cell with the same dimensions as the distances between square faces of the

octahedron, will produce a perfectly filled, cubic simulation cell (I will

use this for FFT computations to get electrostatic energies for individual

atoms).

Dave

-----------------------------------------------------------------------

The AMBER Mail Reflector

To post, send mail to amber.scripps.edu

To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu

Received on Wed Aug 22 2007 - 06:07:48 PDT

Date: Tue, 21 Aug 2007 17:07:32 -0700 (PDT)

Update: it seems I have found the solution ot my problem. If anyone is

interested, the Octahedral box in AMBER is rotated away from the

orientation I was hoping for, which would be to have square faces on the

XY/XZ/YZ planes. To get it back into this orientation, one must apply the

transformation matrix:

cos1 = cos(PI/4.0);

sin1 = sin(PI/4.0);

cos2 = sqrt(2.0)/sqrt(3.0);

sin2 = 1.0/sqrt(3.0);

R = [

cos1*cos2 -cos1*sin2 sin1

-sin1*cos2 sin2*sin1 cos1

-sin2 -cos2 0.0

];

to the coordinates in the AMBER coordinates file. (This is the reverse

rotation as can be found in the ptraj source code actions.c.)

Then, the dimensions given in the AMBER coordinates file apply to the

distance between two hexagonal faces. Multiply those by 2.0/sqrt(3.0) to

obtain the distance between two square faces. Then, all you need to do to

tile the box is translate the unit cell to different lattice positions:

1 0 0

0 1 0

0 0 1

-1 0 0

0 -1 0

0 0 -1

1/2 1/2 1/2

1/2 -1/2 1/2

-1/2 1/2 1/2

-1/2 -1/2 1/2

1/2 1/2 -1/2

1/2 -1/2 -1/2

-1/2 1/2 -1/2

-1/2 -1/2 -1/2

Note that replicating the octahedron at one of the (1/2 1/2 1/2) lattice

positions, then re-imaging all particles as if there were an orthonormal

cell with the same dimensions as the distances between square faces of the

octahedron, will produce a perfectly filled, cubic simulation cell (I will

use this for FFT computations to get electrostatic energies for individual

atoms).

Dave

-----------------------------------------------------------------------

The AMBER Mail Reflector

To post, send mail to amber.scripps.edu

To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu

Received on Wed Aug 22 2007 - 06:07:48 PDT

Custom Search