AMBER Archive (2006)

Subject: RE: AMBER: potential energy

From: Ross Walker (
Date: Mon Sep 18 2006 - 10:52:23 CDT

Dear Fenghui,

> For the results of MD,there are kinetic energy,
> potential energy and relative energy. I think all
> these should be relative quantity. Will you please
> tell me in which state the protein has absolute 0
> energy?

The kinetic energy zero will be when the protein is at absolute zero kelvin.
In other words when all the velocities are zero (or one is doing

The potential energy zero is not within the phase space available to the
protein. (I believe. It would be interesting to see if anyone can prove this
mathematically for a given size protein). Essentially the zero point is when
all of the bonds, angles and dihedrals are at their equilibrium geometries
and the electrostatics and van der waals terms exactly balance. Suffice to
say that unlike quantum mechanical calculations where all molecules /
simulations share the same potential energy origin: All atoms seperated to
infinity, this is not the case with classical MD calculations. Here the
origin is different for different molecules / simulations. Hence you can
only compare delta differences in relative energies.
> After energy minimization, we get a protein
> conformation with the lowest energy and this number is
> usually negative. Can we difine the energy in this
> conformation as 0 energy?

As above the origin is a completely arbritrary position and is different for
each molecule. So technically you can call any point you like the
confirmation with zero energy. Note unless you know you are at the global
minimum (which is unlikely for a protein since finding this is an NPComplete
problem) then there is always the possibility that you could find structures
of lower energy than what you have at the moment during your simulation.
Minimisation has simply found you the nearest low energy minimum.

> For the kinetics energy, it
> should has some relation with the velocity, however
> the initial velocity is produced randomly. Does this
> means that the initial kinetic energy will be a random
> value?

No, the kinetic energy is a direct function of the temperature. Thus you set
what you want the initial temperature to be and velocities are assigned
randomly (weighted by a Boltzmann distribution) to the atoms such that the
total kinetic energy results in the temperature you requested.

You may want to refer to some graduate level texts on computational
chemistry and statistical mechanics to make sure you understand the theory
before attempting any major calculations.

All the best

|\oss Walker

| HPC Consultant and Staff Scientist |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- |
| | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to