VMD-L Mailing List

From: Jan Saam (jan.saam_at_gmail.com)
Date: Fri May 16 2008 - 18:15:51 CDT

Dear Cong Chen,

the reason why you don't see any change after the transformation was
applied is probably because the transformation matrix is the identity
matrix. That means no periodic cell was defined. You can query this with
the
molinfo command using the alpha, beta, gamma, a, b, and c keywords.
Example:
molinfo top get a
gives you the length of the first unit cell vector in Angstroms (for the
current frame).

When all lengths are zero then there are no periodic boundaries defined.

If your DCD file contains the unitcell info VMD will automatically set
the values accordingly. NAMD writes this info into the DCD only if you
specify the option
DCDUnitCell yes
Note that PDB/PSF files do not contain periodic cell information. If you
are dealing with these files you have to set the unit cell manually
using the molinfo commands.

Good luck,
Jan

> ----- Forwarded message from cong chen <vmd_chcwaaa_at_yahoo.com.cn> -----
>
> Date: Fri, 9 May 2008 09:53:08 +0800 (CST)
> From: cong chen <vmd_chcwaaa_at_yahoo.com.cn>
> Subject: vmd-l: about measure dihedral angle
> To: vmd-l_at_ks.uiuc.edu
>
> Dear all,
> I tried to read the source code. To calculate a dihedral angle, the subroutine calculate_dihed has been used. And a subroutine normal_atom_coord has been applied to get coordinates. In this subroutine, there are several codes that I am quite puzzled by,
> ////////////////////////////////////////////////////////////////////////////////////////////////////////////////
> // Apply periodic image transformation before returning
> Matrix4 mat;
> now->get_transform_from_cell(cell, mat);
> mat.multpoint3d(pos, pos);
> ///////////////////////////////////////////////////////////////////////////////////////////////////////////////
> The author said that it is to apply periodic image transformation before returning, but I traced the code step by step and found that the array "pos" has not been changed anymore by using mat.multpoint3d.
> I am quite puzzled. VMD processes NAMD outputs which contain the true positions of the atoms. Why are the coordinates not changed here ? For calculating RDF, the coordinates have been changed if periodic boundary condions have been applied.
> I would be most appreciative if some could give me any advice.
>
>
>
>
> ---------------------------------
> ???????????????????????
> ----- End forwarded message -----
>
>