On Sat, Jun 4, 2016 at 7:14 AM, Himanshu Joshi <himanshuphy87.gmail.com> wrote:
> Hello folks,
> I am following the tutorial by Cheatham et al at AMBER15 work shop at
> London to perform PCA of a structure in MD simulation.
> http://www.amber.utah.edu/AMBER-workshop/London-2015/pca/
>
> I extracted the eigenmode data file from the covariance matrix of
> coordinates followed by the projection of the structure along the few
> principle components (PC) using the "projection" command as described in
> the in above tutorial. If I understand it correctly, it is the dot product
> of the eigen vector and the coordinates of this frame in Angstrom. Please
> correct me if it is not true.
It's true.
> What does it means that the coordinates
> "match well" with the PC. What the relatively high projection to one
> particular mode reflects, some description would be really helpful.
I believe you are referring to this sentence:
'We can now project the trajectory coordinates along PCs to see how
much the coordinates of each frame "match up" along each principal
component.'
Perhaps not the best description - all this means is that when you
project coordinates along an eigenvector you're getting a sense for
how far each set of coordinates (i.e. each coordinate frame) has
"traveled along" that eigenvector (maybe that's not the best
explanation either). A "high" projection just means the coordinates
have traveled quite far along that eigenvector. Remember, the
eigenvectors are just defining new coordinate axes for the movement of
atoms. They don't necessarily describe actual paths the atoms take
during the simulation.
>
> My other question is how do I know how much does the corresponding PC
> contribute to the total motion.
Check out the 'modes eigenval' analysis command in cpptraj.
-Dan
>
> Thank you in advance.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 307
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Jun 05 2016 - 11:00:02 PDT
