AMBER Archive (2008)

Subject: Re: AMBER: about mmpbsa out data

From: steinbrt_at_rci.rutgers.edu
Date: Tue Dec 23 2008 - 22:53:02 CST

Hi, Alice,

> the PBTOT for "Complex", "Receptor", "ligand" seems abnormal: Large STD
> and positive PBTOT for receptor and ligand.
> So, I'm wander is this out data believable?
> Thanks in advance!!!

What you see is not bad per se. The absolute energy of a molecule as
calculated by a forcefield is not a meaningful quantity in itself. You
need to construct a thermodynamic cycle in which all the non-physical
terms cancel out, and taking the difference between the COM and LIG+REC
values in MM-PBSA is one way to do that.

Note that the PBCAL part of both the receptor and complex are fairly
negative, as they should be, since PBCAL should give you the (in principle
physically meaningful if hardly measureable) absolute solvation free
energy of the molecule.

In contrast, the internal energy of the protein is given by the forcefield
terms which are only designed to give relative conformational energies,
not absolute ones.

The large stddev for the INT terms is more of a concern and it is
cancelling out suspiciously well in the final difference. Without more
knowledge of your system its hard to judge whats going on.

Just one guess: Is it possible that your simulation had a disulphide
bridge that you forgot to put into the waterless prmtops for MM-PBSA? This
would explain the large fluctuations but cancel in your end result.

Note that you get a fairly reasonyble total deltaG of -16 kcal, so maybe
your doing fine...

Regards and Happy Holidays,

Dr. Thomas Steinbrecher
BioMaps Institute
Rutgers University
610 Taylor Rd.
Piscataway, NJ 08854
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu