[gmx-users] PMF and Histograms

[Justin Lemkul]

On 1/28/13 9:00 AM, Steven Neumann wrote:
> Dear Gmx Users, Dear Justin,
>
> I run umbrella samplig with small molecule binding protein. With 11
> windows and 0.2 nm spacing I got:
>
> PMF:
>
> http://speedy.sh/zevcp/profile.JPG
>
> Hisotogram:
>
> http://speedy.sh/8Cua2/histo.JPG
>
> (I used -min and -max options of g_wham so leftmost and rightmost
> hisotgrams are cut)
>
> Each window is 100 ns. Could you please advise me why my PMF curve is
> not smooth? The result is something I expect but do not understand the
> little sharp edges on my plot.
> How can I make it smooth?
>

You can't force your data to fit a preconceived notion of what is aesthetically 
pleasing.  Look at the value of DeltaG - it's less than 1 kcal/mol.  Each of 
those fluctuations indicates an extremely small value of energy.  I don't see 
anything wrong with the results.  g_wham has the ability to produce very nice 
error estimates; if you have doubts about your convergence or reliability of the 
outcome, I suggest you investigate those features.

-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

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