AMBER Archive (2009)

Subject: Re: [AMBER] wrong estimation of hydrogen velocities when using SHAKE?

From: Jose Borreguero (
Date: Wed Mar 11 2009 - 16:56:03 CDT

Dear Thomas,

You are right. I looked at several H-N groups in the backbone. Using
Ek(N)+Ek(H)=(5/2)kT result in T's around the system temperature, with
deviations below 2%, which I think is acceptable.


On Wed, Mar 11, 2009 at 2:56 PM, <> wrote:

> Dear Jose,
> > Your assumption is correct: (3/2)kT for the X atom and (2/2)kT for the H
> > atom makes a total of (5/2)kT for the X-H system.
> > The kinetic energy of the X-H system is still the sum of the kinetic
> > energies of each atom, no matter which constraints you decide to impose.
> what I was trying to say is, who tells you that it's the hydrogen that is
> constrained? I could also imagine the SHAKE algorithms to constrain the
> connected heavy atom's position with respect to that of the freely moving
> hydrogen, then it would be 3/2 kT for H and 2/2 kT for X. The reality will
> be somewhere in between, with both atoms loosing some kinetic energy. The
> sum should always be 5/2 kT, but it is not clear how it is distributed
> between the atoms.
> >> > I got T about 6% higher than system temperature. Pretty close.
> My prediction would be that you will find the 'extra' kinetic energy of
> the hydrogens missing from the respective heavy atoms.
> Kind Regards,
> Thomas
> Dr. Thomas Steinbrecher
> BioMaps Institute
> Rutgers University
> 610 Taylor Rd.
> Piscataway, NJ 08854
> _______________________________________________
> AMBER mailing list
AMBER mailing list