[gmx-users] principal axis of inertia - gmx principal

[edesantis]

dear gromacs users,

I'd like to study the shape of some protein aggregates formed during my 
simulation.

In order to do it, I thought to use gmx principal tool to calculate the 
momentum of inertia of the aggregate.

I've never used that tool, so I started studying very simple shapes on a 
single snapshot of the simulation, just to see if I've well understood 
the theory.

Using a script I've generated two selection of atoms, the first is a 
sphere, the second is a  cylinder (its bases is much smaller then its 
height) and then I performed the analysis using the following command

gmx principal -f $gro_file -s $tpr_file -n $index_file -a1 a1.xvg -a2 
a2.xvg -a3 a3.xvg -om princ_ax.xvg



As I understood, the princ_ax file should contain the vectors of the 
principal axes of inertia.
For the case of the sphere, I obtain in the output similar values, this 
means the principal axes of a sphere are of similar magnitude (as 
expected).

For the case of the cylinder, I get a bit weird results, there are two 
values very similar and a third one that is different and much smaller 
than the other two.
I would have expected to have, on the contrary, the two similar values 
to be much smaller then the third, since the dimension in the height of 
the cylinder is more than its bases.


what am I missing?


thank you in advance,
best regards,
Emiliano