From: Sachin Natesh (snatesh_at_uchicago.edu)
Date: Thu Jun 30 2016 - 20:36:07 CDT

Hello all,

I am trying to compute the diffusion coefficient of water molecules that
begin inside a porous cavity of a protein fibril. These waters leave the
unit cell rather quickly in the unwrapped trajectory, and I need to use
these unfolded coordinates for computing MSD. As of now, I do a brute
force, geometric search over the first few thousand cells for water in
these cavities and proceed with my calculation. However, it would be much
easier and more accurate if I just corrected for pbc crossings in the
wrapped trajectory. Does anyone know of a way to write out image flags
indicating how many times an atom has crossed a pbc in some direction given
the dcd trajectory file, or alternative means to accomplish these
corrections? I think it's feasible to record the min/max simulation box
coordinates at each frame and compare atom coordinates to these to detect
pbc crossings, though flags would make things much easier. Any help is
appreciated!

Thanks and regards,
Sachin Natesh