It would be useful to know more about your calculation. However, it is important to recognize that APBS only calculates a portion of the overall binding energy: the electrostatic (polar solvation) contribution and, depending on how you configure the calculation, the apolar solvation contribution. Conformational changes can contribute important entropic and enthalpic terms to the energy which are not calculated by APBS.
Implicit solvent calculation methods such as APBS are often most useful in calculating relative free energy differences (e.g., changes upon mutation) if you are willing to assume that some of the free energy components don’t change very much from mutant to mutant.
I hope this helps,
Pacific Northwest National Laboratory
From: Javier Cáceres [mailto:firstname.lastname@example.org]
Sent: Friday, August 30, 2013 9:58 AM
Subject: [apbs-users] Problem with getting positive binding energies
Hello APBS users,
Im trying to calculate binding energies for a protein-protein complex. Until now, my calc only gave me positive values, which is contradictory because that means that energy desetabilizes the
complex formation. I read and some papers say that the dielectric boundaries are important for de sign of the energy. I use srad values of 0.00 and 1.4, but with this I only got changes in the magnitude but not in the sign.
I can't find what Im doing wrong. I attached the input file in case you want to see.
Thanks and I'll wait for you help.