Bond-jitter/Total-energy-spikes during energy minimization in vacuum

From: Eric Smoll (ericsmoll_at_gmail.com)
Date: Mon May 01 2017 - 02:58:10 CDT

Hello NAMD users,

I have experience with other MD programs but I am new to NAMD. I am trying
to minimize an isolated (gas-phase) small molecule ion. I am using an
OPLS-style force-field. I am using the most recent version of NAMD from
github.

The 2500 step minimization runs without error. When I inspect the
trajectory and the total energy with VMD, It is clear that the molecule is
relaxing: the conformation changes and the overall energy drops. However,
the minimization trajectory shows a large number single-frame distortions.
When viewing the total energy profile, these distortions appear as spikes
on the overall minimization envelope. When viewing the molecular structure,
it appears as if the molecule undergoes a small distortion (a change in
bond-length, for example) before relaxing back to its equilibrium position.
These random distortions persist to the end of the minimization regardless
of the duration (100, 1000, 10000 steps). When the minimization decay
envelope is nearly flat, the spikes become perfectly regular in amplitude
and interval.

Does anyone know what I am doing wrong?

Best,
Eric

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:15 CST