Hi Jason,
Thanks for your reply. The nonpolar solvation energies are -3.4672 and -31.1906 kcal/mol using " &pb istrng=0.1," in AmberTools 1.5 and " &pb istrng=0.1, sander_apbs=1,radiopt=0", respectively.
Does it mean that the nopolar solvation energies are both calculated by the equation G(np) = surften*SASA+offset in the above two cases according to your email? Thanks!
Best wishes
Duan Baogen
From: Jason Swails
Date: 2013-04-10 05:56
To: dbaogen; AMBER Mailing List
Subject: Re: [AMBER] large difference of nonpolar solvation free energy using MMPBSA.py in AmberTools1.5 or AmberTools12
On Tue, Apr 9, 2013 at 3:24 AM, dbaogen <dbaogen.gmail.com> wrote:
Dear all,
I have met a problem in calculation binding free energy using MMPBSA.py script in two versions of AmberTools. one is Amber 11 combined with AmberTools 1.5. The nonpolar solvation energy (ECAVITY ) is -3.4672 kcal/mol. And the input file is as follow:
MMPBSA input file for Amber 11 and AmberTools 1.5
&general
startframe=1, endframe=100, interval=1,
keep_files=0,strip_mdcrd=0,verbose=1,
/
&pb
istrng=0.1,
/
Searching in the AmberTools 1.5 Manual, the nonpolar solvation energy is calculated by using the equation G(np) = SurfTen*SASA+SurOFF = G(cavity) under the above parameter settings.
The other is Amber 11 and AmberTools 12. The input file:
MMPBSA input file for Amber 11 and AmberTools 12
&general
startframe=1, endframe=100, interval=1,
keep_files=2,search_path=1,verbose=1,
/
&pb
istrng=0.1, sander_apbs=1,radiopt=0
/
In this condition, according to the manual, the nonpolar solvation free energy is calculated by using G(np)=G(disp)+G(cavity), and the value (ENPOLAR) is -31.1906 kcal/mol. In addition, the AmberTools 12 manual mentions that the default parameters must be INP=2 with radiopt =1. whether did it mean that I set the wrong parameters for MMPBSA.py script?
This is not right, I don't think. The dispersion/cavitation decomposition of the non-polar solvation free energy is only available via the PBSA source code, _not_ via sander.APBS. Therefore, I think that the surface area term here is the simple term G(np) = surften*SASA+offset.
If I changed the "radiopt" to 1, the error message is "CalcError: sander.APBS failed with prmtop" when sander_apbs is set to 1. If the parameter radiopt=1 and sander_apbs=0, the error message is " PB Bomb in pb_aaradi(): No radius assigned for atom 3 C5' CI ". The program is normally ended only when radiopt is set to 0.
You must either add the missing atom types to the PBSA source code and recompile, or settle with the inp=1 model for non-polar solvation free energy term.
Would you like to give me some help about large difference of nonpolar solvation energy and radii selection of nonpolar energy calculation? Thanks in advance!
I didn't see any numbers, and have no idea what differences you are referring to. Note that inp=2 requires AmberTools to be fully patched (at least up to bugfix.27) in order to get the correct values for the dispersion energy.
All the best,
Jason
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Apr 09 2013 - 19:00:02 PDT
