On Mon, 2014-04-28 at 10:58 +0400, James Starlight wrote:
> Jason,
>
>
> many thanks for explanations again! Some error have been occurred
> during PB with the inclusion of entropy calculations- according to the
> below log It seems that problems have been associated with the
> quasi-harmonic analysis (This run have been initiated on 4 cores).
>
>
> Beginning PB calculations
> with /home/own/Documents/simulations/amber/amber14/bin/sander
> calculating complex contribution...
> calculating receptor contribution...
> calculating receptor contribution...
> calculating receptor contribution...
> calculating receptor contribution...
> calculating ligand contribution...
> calculating ligand contribution...
> calculating ligand contribution...
> calculating ligand contribution...
It looks like you ran in parallel incorrectly. You must use the
MMPBSA.py.MPI executable instead of MMPBSA.py if you want to run in
parallel. This looks like you did
mpirun -np 4 MMPBSA.py -i ...
In this case you are running 4 copies of the same simulation over all
cores, each of which is simulating all frames and overwriting the output
files of each other.
> Does it possible to make some corrections in the input data and
> re-start calculations from the unfinished step not performing long PB
> calculations from the beginning ?
No.
>
> By the way from the general assumption the main entropy contribution
> to the total binding entropy should be associated with the solvent
> entropy which has been implicitly included in the SASA term.
The solvent entropy has been implicitly included in the entire GB
equation, not just the SASA term. The GB equation gives a free energy
-- backing the entropy contribution out of the expression is non-trivial
(or at least non-trivial to me).
> How such solute's vibrational entropy could contribute to the total
> dG?
If this answer was known there would be no point in computing it -- you
could just add it as a post-hoc correction.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Apr 28 2014 - 05:00:04 PDT
