[AMBER] lifetime analysis of native/nonnative contacts data

From: Ruth Helena Tichauer <rhtichau.laas.fr>
Date: Fri, 1 Jun 2018 13:28:46 +0200

Dear Amber users and developers,

After computing native/nonnative contacts, I’ve run lifetime analysis of the obtained native/nonnative contacts data.
I have several questions related to the meaning and interpretation of the obtained curves.

1. If I understand correct, lifetime analysis provides the probability of a native/nonnative contact being present at a given time. Is that correct?

2. Yet, the number of frames of lifetime curves do not correspond to the total number of the trajectory frames but to the longest lifetime encountered, is it right? So if we have a trajectory of 500 frames and the longest native/nonnative contact is present in 200 frames in a row, the corresponding lifetime curve will have 200 frames. Is this right?

3. Sorry for the very unexperienced question but I wonder then how to interpret the lifetime probability. Again, if we have a trajectory of 500 frames and a native/nonnative contact is present the 500 frames, then the lifetime equals 1 and that means that the contact is present all of the time. But, as in most cases, if a given strong native/nonnative contact has several lifetimes how to interpret the value of the lifetime at a given frame? For a trajectory 20 frames long, if the lifetime curve value equals 8 at frame 5, 4 at frame 10 and 0 at frame 15 for example, do we say then that the probability of this contact being present decays by half every 1/4 of the total trajectory?

4. If a nonnative contact appears at the half of a trajectory and remains until the end, can this feature be depicted by this sort of analysis? If yes, how?

5. Is it more suitable to plot normalised or raw lifetime curves for presenting the results? And how these curves are normalised (all values divided by the maximum lifetime curve value)?

Thank you in advance for any insight on this matter and eventually for any book/website that could help me better understand the basis of this analysis.


AMBER mailing list
Received on Fri Jun 01 2018 - 04:30:02 PDT
Custom Search