[gmx-users] Number of bins for wham and umbrella sampling / PMF

[Andreas Mecklenfeld]

Dear Gromacs users,

I want to perform umbrella sampling in order to calculate the PMF for 
pulling a molecule from a surface. For evaluation, I use gmx wham with 
bootstrapping, and Gromacs version is 2018.4.
I've noticed that my results strongly depend on the number of bins, with 
peak heights differ in the magnitude of several kJ/mol. The histograms 
display reasonable overlap. I've tried to alter the number of bins to 
search for convergence (starting from default 200, I used up to 10000 
bins), though convergence was not achieved. Is this an expected 
behaviour and is the default 200 a suitable number of bins?

Thanks,
Andreas


-- 
M. Sc. Andreas Mecklenfeld
Technische Universität Braunschweig
Institut für Thermodynamik
Hans-Sommer-Straße 5
38106 Braunschweig
Deutschland / Germany

Tel: +49 (0)531 391-2634
      +49 (0)531 391-65685
Fax: +49 (0)531 391-7814

http://www.ift-bs.de