From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Wed Aug 11 2010 - 08:44:03 CDT
On Wed, Aug 11, 2010 at 8:48 AM, Mert Gür <gurmert_at_gmail.com> wrote:
> Dear Axel,
> The force field is an equation right?
> If you have an equation which depends on some parameters (which are
> the atom coordinates in MD) there are some specific values of
> parameters that set this equation to zero.
> I don't know how else you would call that structure that has these
> specific atomic coordinates and an 0 energy value.
your answer is the philosophically correct answer.
my answer is the practical one.
if i follow your reasoning, the reference that you are referring
to would be 0 atoms or all force field terms set to zero. if you
ask why is the total energy positive or negative _at all_, your
answer is the correct one. ultimately, the absolute energy value
is completely irrelevant (particularly since it is missing all the
energy terms that have been "integrated out" due to doing
classical mechanics instead of more detailed descriptions)
and can be arbitrarily shifted.
it can be however, very misleading in the context the question
was asked. which i would paraphrase as: why do some of my
proteins have a positive total energy and some a negative
total energy when i remove the solvent and the total energy
with the solvent is negative? the answer to that is, of course,
that strongly attractive solvent-solute interactions can overcompensate
repulsive interactions within the protein.
hope that clears matters up,
> I would call it reference state since Namd computes the energy values
> (positive or negative) with respect to that state.
> Imagine my point of view as follows
> If the force field would have been an harmonic function the reference
> state would be the native conformation and this is set to zero in NMA.
> On Wed, Aug 11, 2010 at 12:35 PM, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>> On Wed, Aug 11, 2010 at 4:15 AM, Mert Gür <gurmert_at_gmail.com> wrote:
>>> The energies are evaluated with respect to an reference state
>>> (reference energy). Therefore it is possible that you observe both
>>> positive an negative energies.
>> what do you mean by "reference state". namdenergy does not
>> compare the result to anything, it just creates in input file for namd,
>> writes out the selected atoms to a matching coordinate and psf file,
>> runs namd and parses the result. as such the interactions are
>> computed directly according to the force field from the force field
>> now, if you take away the protein solvent interactions, that may
>> result in a significant change of the total energy, particularly if
>> the protein has overall strong interactions with the solvent.
>> also, for a single snapshot taken arbitrarily from an ongoing
>> simulation, it is quite possible, that the protein conformation
>> for away from its ground state and thus of higher potential energy.
>>> On Tue, Aug 10, 2010 at 4:48 PM, C. Navarro <kdefuego_at_gmail.com> wrote:
>>>> Hello all,
>>>> I'm having some trouble trying to understand if it can be OK to have
>>>> positive energy for a protein during molecular dynamics with namd. I'm
>>>> afraid my phisics and termodinamics skills are not enough to answer this
>>>> question by myself, so I'm putting it to this list.
>>>> After 1 ns dynamics of a cytosolic protein (in water), I have taken the
>>>> dcd file and opened in vmd, then calculated the intramolecular energy of
>>>> the protein for every frame, with this tcl commands:
>>>> set sel [atomselect top "protein"]
>>>> package require namdenergy
>>>> namdenergy -all -sel $sel -ofile protein-energy.dat
>>>> Giving me positive total energies. It is important to say that this is
>>>> happening only with one copy of three? of my protein of study, (The
>>>> other two proteins show negative energy values); so it is hard to me to
>>>> think in an error during the dynamics protocol. By the way, the total
>>>> energy of the system (taking all the atoms for the energy evaluation),
>>>> still gives negative values (which is fine).
>>>> Please, any input to this issue would be very appreciated.
>> Dr. Axel Kohlmeyer akohlmey_at_gmail.com
>> Institute for Computational Molecular Science
>> Temple University, Philadelphia PA, USA.
-- Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://sites.google.com/site/akohlmey/ Institute for Computational Molecular Science Temple University, Philadelphia PA, USA.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:24 CST