[gmx-users] Representative structure from MD trajectories

[bipin singh]

Hello,

I have constructed a 2-d FEL using PC1 and PC2 as the reaction
coordinates(by using g_sham).
>From this 2-d FEL, I extracted the PC1 and PC2 (principal components)
values corresponding to minimum free energy.
Then I extracted the time during which system possess these PC's
values( by looking at PC Vs time plot).
Now I want to extract representative structures from these time
length(for eg. 40ns to 70 ns) of MD trajectories.


This question had been discussed previously but I am not able to find
a consensus solution:

There are three approaches discussed in the previous posts:
(1)Use g_cluster with options -av(writes average) -cl and then
minimize the structure
(2)Use g_covar -av, and then minimize the average structure.
(3)Use g_rmsf -ox, and minimize the structure


Please suggest which approach should be used for this purpose.




-- 
-----------------------
Thanks and Regards,
Bipin Singh