Re: Pair interaction energy

From: Brian Bennion (brian_at_youkai.llnl.gov)
Date: Fri Jan 30 2004 - 12:38:35 CST

On Fri, 30 Jan 2004, Ioana Cozmuta wrote:

> > > On Fri, 30 Jan 2004, Sundaramoorthy, Munirathinam wrote:
> > >
> > > > Dear NAMD users,
> > > >
> > > > Alternatively, is it good enough if I calculate energies of individual
> > > > chains (A & B) and the dimer (AB) and use the difference [(A+B)-AB] as
> > > > the interaction energy?
> > > >
> > > Just a comment: you would have to subtract big numbers so you'll have
> > > large uncertainty on your result.
> > >
> > > Ioana
> > However, you could decrease your error somewhat by using a few MD runs to
> > shake the interface around and then make a calculation. If you follow the
> > logic in the first email, you would of course have to run MD on the
> > dimers. Although I would not expect large changes in the potential
> > energy.....
> >
>
> here;s an example. Say you calculate an energy value for individual chain
> A of -3500kcal/mol and for B -3600 kcal/mol. For the dimer you get -7105
> kcal/mol. The interaction energy would be then according to the formula
> above -5 kcal/mol. Normally you do not have uncertainties of less than 5%
> in MD simulations (specially if you use different values for the seed). So
> the -5 kcal/mol value that you get by subtracting the two is not
> significant within the error bars...

Assuming your values and error assumptions are realistic then yes your
reasoning follows.
However, it should still be tested. Well planned MD simulations can get
you a long way on reducing error. I just finished reading a paper where
the authors proposed an accuracy within kbT. Don't know if I believe it
yet, but it was interesting.

Brian

>
> Ioana
>

*****************************************************************
**Brian Bennion, Ph.D. **
**Computational and Systems Biology Division **
**Biology and Biotechnology Research Program **
**Lawrence Livermore National Laboratory **
**P.O. Box 808, L-448 bennion1_at_llnl.gov **
**7000 East Avenue phone: (925) 422-5722 **
**Livermore, CA 94550 fax: (925) 424-6605 **
*****************************************************************

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:38:23 CST