# Re: Truncated octahedron box issues (?)

From: Jim Phillips (jim_at_ks.uiuc.edu)
Date: Thu May 26 2011 - 22:59:06 CDT

Given the symmetry of the lattice you can probably change any of the signs
without affecting the Weigner-Seitz primitive cell (i.e., wrapNearest).

-Jim

On Thu, 26 May 2011, Massimiliano Porrini wrote:

> Dear Nicholas and Jim,
>
> Thank you for your useful replies.
>
> Yesterday I tried with the basis vectors given in the link suggested
> by Nicholas and
> it worked.
>
> Today I tried also the ones given by Jim and it worked as well.
>
> But the two matrices differ by the sign of some elements (as you can see)
>
> Does anybody know if there is a way to demonstrate that the two matrices are
> equivalent?
>
> Am I on the safe side if I choose one or the other?
>
> Many thanks in advance,
> MP
>
>
> 2011/5/25 Jim Phillips <jim_at_ks.uiuc.edu>:
>>
>> I think you have the matrix transposed.  Try it like this:
>>
>> cellBasisVector1 \$xx \$yx \$zx
>> cellBasisVector2 \$xy \$yy \$zy
>> cellBasisVector3 \$xz \$yz \$zz
>>
>> Try loading the original structure into VMD and using the draw line command
>> as in the code below to draw your original and swapped basis vectors.  The
>> swapped (green) basis vectors should be orthogonal to the facets of the
>> original cell.
>>
>> -Jim
>>
>>
>>
>> set dist 58.9627607
>>
>> set xx [ expr \$dist ]
>> set xy [ expr \$dist * 1/3 ]
>> set xz [ expr \$dist * -1/3 ]
>>
>> set yx [ expr 0 ]
>> set yy [ expr \$dist * sqrt(2) * 2/3 ]
>> set yz [ expr \$dist * sqrt(2) * 1/3 ]
>>
>> set zx [ expr 0 ]
>> set zy [ expr 0 ]
>> set zz [ expr \$dist * sqrt(6) * 1/3 ]
>>
>> set o [list 29.295766830444336 29.650468826293945 29.59993553161621]
>>
>> set a [list \$xx \$xy \$xz]
>> set b [list \$yx \$yy \$yz]
>> set c [list \$zx \$zy \$zz]
>>
>> draw color white
>> draw line \$o [vecadd \$o \$a]
>> draw line \$o [vecadd \$o \$b]
>> draw line \$o [vecadd \$o \$c]
>> draw line \$o [vecsub \$o \$a]
>> draw line \$o [vecsub \$o \$b]
>> draw line \$o [vecsub \$o \$c]
>>
>> puts [veclength \$a]
>> puts [veclength \$b]
>> puts [veclength \$c]
>>
>> set a2 [list \$xx \$yx \$zx]
>> set b2 [list \$xy \$yy \$zy]
>> set c2 [list \$xz \$yz \$zz]
>>
>> draw color green
>> draw line \$o [vecadd \$o \$a2]
>> draw line \$o [vecadd \$o \$b2]
>> draw line \$o [vecadd \$o \$c2]
>> draw line \$o [vecsub \$o \$a2]
>> draw line \$o [vecsub \$o \$b2]
>> draw line \$o [vecsub \$o \$c2]
>>
>> puts [veclength \$a2]
>> puts [veclength \$b2]
>> puts [veclength \$c2]
>>
>>
>>
>> On Wed, 25 May 2011, Massimiliano Porrini wrote:
>>
>>> Dear all,
>>>
>>> I am trying to simulate a small protein immersed in a truncated
>>> octahedron box of water molecules.
>>>
>>> Basically I built the box via tleap program of Amber10 and then (using
>>> the vectors component
>>> taken from Gromacs manual) I passed the box parameters to NAMD config
>>> files in the
>>> following way (the "cellOrigin" values were obtained via VMD):
>>>
>>>
>>> 1) MINIMIZATION:
>>>
>>> ***
>>> # Periodic Boundary Conditions
>>> set dist 58.9627607
>>>
>>> set xx [ expr \$dist ]
>>> set xy [ expr \$dist * 1/3 ]
>>> set xz [ expr \$dist * -1/3 ]
>>>
>>> set yx [ expr 0 ]
>>> set yy [ expr \$dist * sqrt(2) * 2/3 ]
>>> set yz [ expr \$dist * sqrt(2) * 1/3 ]
>>>
>>> set zx [ expr 0 ]
>>> set zy [ expr 0 ]
>>> set zz [ expr \$dist * sqrt(6) * 1/3 ]
>>>
>>> cellBasisVector1    \$xx   \$xy  \$xz
>>> cellBasisVector2    \$yx   \$yy  \$yz
>>> cellBasisVector3    \$zx   \$zy  \$zz
>>>
>>> cellOrigin          29.295766830444336 29.650468826293945
>>> 29.59993553161621
>>> ***
>>>
>>>
>>> 2) HEATING:
>>>
>>> ***
>>> # Continuing a job from the restart files
>>> binCoordinates     \$inputname.restart.coor
>>> extendedSystem     \$inputname.restart.xsc
>>>
>>> # Restraints
>>> constraints         on
>>> consref             \$filename.pdb
>>> consKfile           \$filename.pdb
>>> consKcol            O
>>>
>>> # Periodic Images
>>> wrapNearest         on
>>> wrapWater           on
>>> wrapAll             on
>>> ***
>>>
>>>
>>> Hence at the end of these two processes I should get exactly the same
>>> box shape and
>>> dimensions as the one created via tleap (I am using an NVT esemble), but
>>> instead, as you can see from the linked figures, the shape is different,
>>> since
>>> the one obtained after NAMD minimization and heating appears to be
>>> "flatter":
>>>
>>> http://www.4shared.com/photo/9kN_g9Qi/amber.html
>>> http://www.4shared.com/photo/CKpojNEP/namd.html
>>>
>>> Do you kindly have any idea/hint about this behaviour?
>>>
>>> Should I be worried about a possible rotation of the protein and thus
>>> a likely interaction with a periodic image of its?
>>>
>>> Many thanks in advance.
>>>
>>> All the best,
>>> MP
>>>
>>> PS: as one can see from the displayed axes in the two enclosed figures,
>>> the orientation of the system is exactly the same, and the "suspected
>>> contraction" occurs along the z axis.
>>>
>>>
>>> --
>>> Dr Massimiliano Porrini
>>> P. E. Barran Research Group
>>> Institute for Condensed Matter and Complex Systems
>>> School of Physics & Astronomy
>>> The University of Edinburgh
>>> James Clerk Maxwell Building
>>> The King's Buildings
>>> Edinburgh EH9 3JZ
>>>
>>> Tel +44-(0)131-650-5229
>>>
>>> E-mails : M.Porrini_at_ed.ac.uk
>>>              mozz76_at_gmail.com
>>>              maxp_at_iesl.forth.gr
>>
>
>
>
> --
> Dr Massimiliano Porrini
> P. E. Barran Research Group
> Institute for Condensed Matter and Complex Systems
> School of Physics & Astronomy
> The University of Edinburgh
> James Clerk Maxwell Building
> The King's Buildings