[gmx-users] Flat energy profile in g_wham

[chris.neale at utoronto.ca]

Alexander:

Can you please also post your free energy profile out of wham? Also,  
can you confirm that you are using different force constants for  
different windows?

Justin is correct, but that just means that you are using more cpus  
than you need... it's not actually wrong. What are your units on the  
x-axis? If they are in nm then it doesn't surprise me at all that you  
have a flat free energy profile.


-- original message --

    Sorry for poor explanations.
My histogram is here http://picasawe b.google.ca/117 558205101329348
732/G_wham#5511 244561053627250
<http://picasaweb.google.ca/117558205101329348732/G_wham#5511244561053627250>
it was 12 ns per window run.
Looks to me like your windows overlap too much - there is complete  
redundancy in
many of them, such that it looks like several of your windows are completely
unnecessary.  See the example histogram in the PMF tutorial for a more  
"normal"
histogram.

-Justin

[Hide Quoted Text]
Regards,
Alex.



      alexander yakovenko wrote:
       > Hi Justin!
       > Thank you for replay. My histogram (can send if required)
      consists of 13
       > windows (0.1 and 0.2nm sampling) and look like two peaks (with
       > amplitudes 3-4e+2) separated with a valley of 1.2-1.5e+2 (my
      protein has
       > two sub-domains so should be OK) and smallest overlap between
      windows is

      I don't understand this description. How can 13 windows produce only
      two peaks?
--