[gmx-users] Fwd: different vectors from gmx covar and gmx anaeig

[Qinghua Liao]

Hello,

I found out what's going on there, the first set data is the total 
vector, which I thought was

the X component.  Thanks!


All the best,

Qinghua



-------- Forwarded Message -------

Subject: 	different vectors from gmx covar and gmx anaeig
Date: 	Thu, 18 Oct 2018 13:30:13 +0200
From: 	Qinghua Liao <scorpio.liao at gmail.com>
To: 	gromacs.org_gmx-users at maillist.sys.kth.se



Hello,

I am doing PCA analysis to characterize the motion of a domain, I got 
the files:

average.pdb
eigenval.xvg
eigenvec.trr
covar.log

Then, with "gmx trjconv -f eigenvec.trr -s average.pdb -o eigenvec.pdb", 
I got all the vectors in the pdb file "eigenvec.pdb"

I also used "gmx anaeig -v eigenvec.trr -eig eigenval.xvg -first 1 -last 
2 -comp eigcomp.xvg " to get the first two vectors component
in "eigcomp.xvg"

I compared the vectors generated with the two commands, and I found that 
they are different.

For example, for the first vector (only the first three atoms are listed),

"eigenvec.pdb":
                                                x          y z
ATOM      1  N   LYS     1       0.090   0.749  -0.555  1.00 
0.00           N
ATOM      2  CA  LYS     1       0.142   0.704  -0.532  1.00 
0.00           C
ATOM      3  C   LYS     1       0.111   0.653  -0.487  1.00 
0.00           C

"eigcomp.xvg":

x
     1.0000    0.09367
     2.0000    0.08940
     3.0000    0.08218
y
     1.0000    0.00905
     2.0000    0.01424
     3.0000    0.01114
z
     1.0000    0.07494
     2.0000    0.07042
     3.0000    0.06526

I guess the unit is nm for both cases. Then how should I interpret this 
differences, did I do something wrong? Thanks a lot!


All the best,
Qinghua