[gmx-users] Re: Dumping of energies

[andersle at stud.ntnu.no]

>> Dear all,
>>
>> I wonder if there is some way in gromacs to dump the potential energy
>> (or internal energy) of each particle to the disk during a simulation.
>
> No, because it isn't calculated.
>
>> That is, something similar to the dumping of the positions, velocities
>> and forces of the individual particles.
>>
>> Is there a way to do this out of the box in gromacs?
>> Or has anyone done this by modifying the source?
>
> Well, first you have to decide how to partition the energy arising from
> each type of bonded interaction onto the participating atoms.
>
> Actually, before that, I'd ask myself what I was hoping this
> per-particle energy was going to tell me.
>
> Mark

For instance:
- thermal conductivity (Green-Kubo relation)
- energy fluctuations in different parts of a inhomogeneous system  
(e.g. protein in water)

Anders