[gmx-users] Help with Principal component analysis

[Mark Abraham]

On 3/01/2012 6:54 AM, Alex Jemulin wrote:
> Dear all
> I run a MD on a GPCR (transmembrane protein)
> Then I run a PCA on results and I found 3PC sufficient to explain 
> variance.
> On the same PC I get big values for samples located both at Nter 
> (extracellular) and Cter (intracellular) or for similar cases such as 
> both Nter(extracellular) and Loop5 (intracellular).According to you 
> how could I explain the motion of group atoms intra and extracellular 
> on the same components?
> Does it mean the motion of some atoms at Nter (for instance) could 
> influence the motion at Cter ? Could it be logic?

Highly unlikely. You don't give any details of your simulation, so your 
simulation could be too short, or have insufficient distance between 
your periodic images.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120103/fe776610/attachment.html>