[gmx-users] EM replicates

[Gregory Poon]

Hello all!

I am wondering if the solvate, genion, and energy minimization (using 
steepest descent) procedures are deterministic.  I have been performing 
replicate energy minimization of a duplex DNA structure, always starting 
from scratch (i.e., pdb2gmx) and using the same parameters in terms of 
force field, water choice, ion concentration/type etc. and getting 
convergent structures that are super-imposable.  My question is whether 
this is something that I would expect in any event, for any system (up 
to energy minimization, before dynamics simulation), or that I just 
happen to have a well-behaved system.

Thanks in advance,
Gregory