From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Fri Oct 06 2017 - 08:15:40 CDT
Vlad's suggestion to use distanceInv is very good, but I would only add a
word of caution not to use very large groups for it, because it loops over
all pairs of atoms, thus the computational cost of such a variable starts
small but can grow very quickly with group size.
Also, distanceInv does not have an implementation of the inverse gradients
and straight ABF will not work with it, but you can switch to eABF,
metadynamics, umbrella sampling, ...
This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:42 CST