On Wed, Mar 06, 2013, Brian Radak wrote:
> As the manual states, if you're just doing minimization to remove contacts,
> it should be ok to just use ntc = ntf = 2. Others might be qualified to
> comment, but I believe that two things the minimizer does not do well with
> are 1) rigid water molecules and 2) periodic boundary conditions. In those
> cases especially, the minimizer can pretty much only be used as a practical
> tool for removing bad contacts.
The minimizer is bad with any rigid (SHAKEN) bonds, not just water. But
periodic boundary conditions should cause no problems. If they do, I think
that is a bug we should address.
Note that NAB can correctly minimize systems with rigid bonds; but on the
other hand, it cannot handle periodic systems. Porting the rigid bond code
into sander has long been on kind of a to-do list, but never got very high on
the list of anyone actually having the time to do it.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Mar 06 2013 - 18:00:03 PST
