[gmx-users] taking snapshots of the protein after simulation

[Mark Abraham]

----- Original Message -----
From: Hassan Shallal <hshallal at PACIFIC.EDU>
Date: Saturday, June 5, 2010 13:16
Subject: [gmx-users] taking snapshots of the protein after simulation
To: gmx-users at gromacs.org


  -----------------------------------------------------------
| 

 
 > Dear Gromacs users, >   > I have had one of my studied proteins simulated in a 5 ns production. > Now, I wonder how can I get, let's say, 100 coordinates files, each file represent the coordinates of the protein at certain time during the simulation, for example, each 50 ps OR.. What if I want these 100 files to represent the most stable and least energetic conformations of the protein during this 5 ns simulation independent of which time they represent... I know that VMD can be used for that but I am not quite sure how I can do it...

As Justin said, trjconv is the tool for you to use to get structures you want, but choosing those structures is a trickier proposition. "Most stable" is an awkward thing to define. "Least energetic" is fine, but you haven't yet got any estimate of free energies... this is why people do clustering, or PCA. "Least potential energy" means much of nothing - this is not ab initio QM on isolated structures. A key idea here is to have read literature about what other people have done with similar objectives in studying systems similar to yours.

This is a good illustration of how planning the manner in which you intend to analyze your results might need to come earlier in your workflow design than after you've collected your data... You might need to do the latter to support the needs of the former.

Mark |
-----------------------------------------------------------



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20100605/c1eb506f/attachment.html>