Using wham with umbrella sampling

[Nawel Mele]

Dear Plumed users,

After performing umbrella sampling simulation using RMSD as CV I would like to plot the PMF and the umbrella histograms using WHAM as in the following  gromacs tuto : http://www.mdtutorials.com/gmx/umbrella/07_analysis.html
but I am not sure how to use wham with plumed, can someone help me here?
I have different US windows simulations with a COLVAR output file for each of them with the restraint.bias values through the simulation window.

Many thanks for your help,

Nawel

[Bogdan Marekha]

Dear Nawel,

Did you have a look at the Belfast Plumed tutorial on umbrella sampling? Part of it is devoted to recovering the unbiased PMF with WHAM.

Best,

Bogdan

[Nawel Mele]

Dear Bogdan,

I did have a look but they used wham when they perform US with different AT values while here I have different windows corresponding to different starting conformations based on the RMSD. In the tutorial they never mentiopnned the used of multiple US windows which makes me a bit confused on how to use the wham methodology.

Nawel

[Bogdan Marekha]

Dear Nawel,

There should be no difference whether you did your simulations consecutively changing the position of the restraint or ran several independent simulations from starting different initial configurations as long as the sampling is adequate.

In the tutorial there is actually a subsection on concatenating the output of the different simulations before perfoming WHAM on the whole dataset.

Best,

Bogdan

[Nawel Mele]

Dear Bogdan,

Many thanks for your quick reply. I was reading again the tutorial and I think I misunderstood something when applying the restraint, is the parameter AT the actual position of the system? I mean, I have extracted frames from another simulations (DIMS simulation going from a active to an inactive form) and I used these frames for my US windows. But should I actually change the AT value for each of them as in the tutorial?

Many thanks again,

Nawel