I have a question about how to interpret the results from doing a
PCA/Quasiharmonic analysis. I am using the following relevant ptraj
commands:
trajin <file>
center
rms first .CA,C,N
matrix mwcovar name mxcvarmat .CA out mwcovarmat.dat
analyze matrix mxcvarmat name evecs out evecs.dat vecs 30
analyze modes displ out Mode1.dat stack evecs beg 1 end 1
...
I then update and do Mode 2 beg 2 end 2, Mode3 beg 3 end 3, etc. My
understanding is that until I get to the 7th mode I am only looking at
translation and rotation, however when I plot the fluctuation profile (delta
r's) for the first six modes the plot looks like physically relevant data.
As I am centering and doing an rms (unless I am not doing it correctly) I
understood that the first six modes should return zero's, which they do not.
Also, looking at the frequencies for the first six eigenvalues they are
continuous with higher values, with non-zero eigenvectors.
I then looked at the last few eigenvectors and they are, in fact, very close
to zero. Is there a sorting that the zero modes are placed at the end, in
which case I should look at the top few modes as physically relevant, or
should I ignore the first six modes as translation and rotation and only
begin looking at Mode 7? Also, why are there no zero values? Shouldn't
there be 6 zero values for translation and rotation?
Thank you very much in advance for any help,
Regards,
Tyler Glembo
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 06 2010 - 23:00:03 PDT
