[gmx-users] Solvation Free Energy calculation (g_bar)

[Daniel]

Hi everyone,

I'm having problems when calculating Free energy through the g_bar tool. I'm
trying to decouple separately the electrostatic and the Vdw terms. So, after
performing the mdruns for all Coulombic processes (from 0 to 1, split in 0.1
steps: 0.0-0.1-0.2...0.9-1.0) and the same for the LJ-terms, I'm applying
the g_bar for all md*.xvg output files, from both decoupling steps, all
together. I mean I join the md*.xvg files from the two decouplings and apply
g_bar only once. As a result I'm getting a warning message informing that
some results violate the Secong law of Thermodynamics and the error
estimation is about 8 KJ/mol!
On the other hand, when I apply the g_bar tool separately for the two
decouplings I no longer "violate the Second Law" and the errors are more
feasible (~0.3 KJ/mol), however I don't know how the join the two results in
order to get the solvation Free energy of the whole process (considering
that this is possible, I don't know this as well).

Any idea?

Daniel.

--
View this message in context: http://gromacs.5086.n6.nabble.com/Solvation-Free-Energy-calculation-g-bar-tp4998662.html
Sent from the GROMACS Users Forum mailing list archive at Nabble.com.