Hi all,
Part way through a long MD simulation of a large protein in a truncated octahedral water box, I've wrapped the protein into the centre of the periodic cell, as has been suggested to me to mitigate some instability in the system.
When I do this in ptraj, by dint of reading a restart file in, performing the wrapping procedure, and writing a restart file out, I find that the velocities are scrapped and replaced with zeroes.
If it were a cubic unit cell, I would feel quite confident in saying that the velocities can just be lifted from the original restart file and put into the new one. But with a truncated octahedron, I'm not sure. My question therefore becomes:
Is it safe to assume that atoms are moving in the same direction before and after the wrapping (so that a simple copy-and-paste can be done with the velocities)? If not, does ptraj wrap the velocities as well as the coordinates, and/or how easy would it be to implement such functionality?
It's not something I'm inclined to lose too much sleep over, since if all else fails I can use the Maxwell-Boltzmann distribution as dished up by PMEMD. However, before I do so (by starting with zero velocities), I thought I'd ask.
Thanks in advance,
Ben
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 28 2010 - 13:30:04 PDT
