I need to find the converged conformational sampling within 200 ns per replica between RMSD and free energy in equal non-overlapping time windows like in the following graph?
@Sam Mohel
RSMD is a rather bad choice of an observable for this matter, because it requires a comparison point. In your case that's probably the starting structure and hence your minimum isn't at 0 deviation, because the best structure isn't the initial one. A better method is dPCA which can be done with various tools, eg. carma (PSF/DCD), depending on your trajectory format, other tools may exist.
I think @Zhaoxi Sun got your plot wrong, as lower is better on a free energy scale.
See our paper on a starter on structural free energy landscapes:
Article Replica Based Protein Structure Sampling Methods: Compromisi...
It's also easily possible, to extract frames of certain free energy and coordinates on the free energy landscape or major structural clusters.
Zhaoxi Sun I see you point. The global minimum is seperated by a energy barrier of about 3 kcal, which might already be a significant change, depending on the system size tough.
- Badges
- Science topic