[gmx-users] umbrella sampling-Bimodal Histograms

[chris.neale at utoronto.ca]

Take that replica on the left that shows a bimodal histogram. Now plot  
a time series of the displacement: x-axis = time and y-axis =  
displacement along reaction coord. Is it jumping back and forth  
between two regions of sampling? Probably not... more likely it starts  
near one maximum and transitions to another maximum.

If this is true then it simply means that you have not equilibrated  
enough. Run longer and cut some of the initial sampling.

Chris.

-- original message --

I am doing US . Yes, histograms of population densities along the reaction
coordinate. Please find the pull settings. Sampled for 800ps.(nsteps =
400000). Pull_init will vary for each frame, depends on the window spacing.

pull                     = umbrella
pull_geometry            = position
pull_dim                 =  N N Y
pull_start               = no
pull_nstxout             =  10
pull_nstfout             =  10
pull_ngroups             =  1
pull_group0              =  U_ref
pull_pbcatom0            = 0
pull_group1              =  r_C1
pull_pbcatom1            = 0
pull_init1               =  0 0 0.1
pull_k1                  =  1000
pull_rate1               = 0
pull_vec1                =  0 0 0

Please check this link for my histograms

https://docs.google.com/fileview?id=0B1PyTWWGrqt6MDU3NWYwMGUtNjY5Zi00NDBmLWE0YzMtYTNjODVlOGFlNWVl&hl=en

I would greatly appreciate our suggestions.

Thank you,
-Aswathy

On Mon, Jun 21, 2010 at 8:14 PM, <chris.neale at utoronto.ca> wrote:

> Please clarify:
>
> Are doing SMD or are you doing US? If you're doing SMD then you should not
> be using WHAM and you should not really be able to generate any sampling
> histograms.
>
> Are the histograms that you are referring to population densities of the
> sampling along your reaction coordinate?
>
> My guess -- if you're doing US -- is that you have some incorrect pull
> group settings. Bimodal distributions are indeed possible, but should
> require very long sampling times to achieve, and I doubt that you are at
> those times yet. Your Fc is fine. Post your pull settings.
>
> Chris.
>
> -- original message --
>
> When i did the Umbrella sampling of frames from an SMD (of ligand
> transport), I am getting bimodal histograms in some cases.
>
> Do you think this is because , the force constant that i used is very low
> (i
> used pull_k1=1000).? Are these bimodal peaks may cause any deviation in my
> PMF result? Do I need to repeat the sampling again with another pull_k1
> value(higher value)?