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

[Justin A. Lemkul]

On 6/18/12 3:05 PM, Daniel wrote:
> Hi Justin,
>
> Thank you for your reply. I see what you mean, but the problem is when I use
> the approach you mentioned (adding the DGs) my result is far from the
> experimental one, whereas when I choose the first one (which violate the
> Second law) my result is quite good, except for the error estimation and the
> warning, It seems that it makes an important difference the way the g_bar is
> applied, considering my simulations are right.
> Now I'm trying to increase the number of lambdas in the region where the
> Second law has been violated in an attempt to solve this issue, Anyway, I'm
> stil not sure if my procedure is correct and why I'm getting such a
> difference when applying the g_bar in these two ways.
>

The warnings you are receiving arise because you're concatenating dissimilar 
data sets (one where LJ are being decoupled, and one where Coulombic 
interactions are being decoupled) and expecting the result to make sense.  It is 
possible to (de)couple both LJ and Coulombic terms simultaneously, but I have 
never done this myself; I have always relied on a two-state (de)coupling procedure.

As to the quality of the result, I can't comment on that because I don't know 
what you're working with.  The quality of the result is dependent upon the 
quality of the topology, so if you're trying to develop new parameters for a 
molecule, you may need to reconsider how the parameters were derived.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================