[gmx-users] g_mindist -pi option

[Miguel Machuqueiro]

At 08:00 AM 6/12/2007, you wrote:
>Hi again,
>since I did not get answer I double checked my trr file. In case I 
>calculate the minimum distance of
>the periodic image for each component of my complex, I get 
>reasonable value of it (3.0 and 2.9 nm),
>whereas if I consider the whole system (protein) I get this strange 
>result of 0.17 nm.
>Any clue?
>thanks again
>Regards

Hi Andrea,

When you have a multi-chain protein (or a protein + peptide) it can 
happen that when making protein whole (removing pbc) sometimes the 
two chains can be placed in two different adjacent boxes, so... when 
doing the "g_mindist -pi" setup, the closest distance is when they 
are together in the same box at a distance of ~1.7 Angstrom.

Point 1 - This is an artifact, but it doesn't exclude the possibility 
that there are structures seeing their periodic images. As it was 
suggested by Mark Abraham, you should check these structures just to be sure.

Point 2 - If someone knows how to trick trjconv -pbc in order to 
avoid doing this... it would be appreciated.

Miguel


========================================
Miguel Machuqueiro
ITQB-Instituto de Tecnologia Química e Biológica
     Universidade Nova de Lisboa
Molecular Simulation Group - 6º Floor (Room 601)
Av. da República, EAN, Apartado 127
2781-901 Oeiras, Portugal
Tel. : +351 214469618 /Mobile: +351 96 7562285
E-mail: machuque at itqb.unl.pt
_________________________________________