[gmx-users] Identifying representative structures from a g_sham generated 2d energy plot

[Bailey A.]

Hi,

Having used g_sham to plot  a Gibbs free energy landscape using a projection of the first two principal components from the normal covariance analysis (not dihedral PCA) like so:

g_sham -f 2dproj_1_2.xvg -ls gibbs.xpm -notime

I would like to now identify which parts of the trajectory are represented by the various parts of the landscape. Are there simple steps to work through to do this?

Having looked through the manual, previous user-list correspondence and output files I can see the bins and their contents and cross reference that with the log and ener.xvg, but I got a bit stuck after that.

I also had a read of the David Leitner paper from 2007 where they identify structures from dPCA generate plot, "By quenching the structures in the space of the first two dPCA eigenvectors", but I wasn't really sure what this means.

Any clues to get me started would be much appreciated.

Alistair
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110916/2e953660/attachment.html>