Hi Ankit,
Chimera does not have a display style to show the inertia of a molecule as 3 axes. But here is a quick way to hack that in. You can change a little bit of Python code in your Chimera 1.10 distribution so that it shows three narrow ellipsoids that represent the axes. Use a text editor to edit
chimera/share/measure/inertia.py
or on Mac Chimera.app/Contents/Resources/share/measure/inertia.py and replace the ellipsoid surface function
def ellipsoid_surface(axes, lengths, center, color, surface):
xf = surface.openState.xform.inverse()
sa, sc = transform_ellipsoid(axes, center, xf)
varray, tarray = ellipsoid_geometry(sc, sa, lengths)
p = surface.addPiece(varray, tarray, color)
p.save_in_session = True
return p
with
def ellipsoid_surface(axes, lengths, center, color, surface):
xf = surface.openState.xform.inverse()
sa, sc = transform_ellipsoid(axes, center, xf)
for sx,sy,sz in [(1,.05,.05), (.05,1,.05), (.05,.05,1)]:
lx,ly,lz = lengths
varray, tarray = ellipsoid_geometry(sc, sa, (lx*sx,ly*sy,lz*sz))
p = surface.addPiece(varray, tarray, color)
p.save_in_session = True
return p
then save the file, restart Chimera and now “measure inertia #0 color yellow” will show the 3 axes. Make sure to keep the indentation of the code since that is important in the Python computer language. This change is scaling down 2 of the 3 ellipsoid axes by a factor of 0.05. You can change the 0.05 value to say 0.01 if you want thinner axis ellipsoids.
Tom
On Feb 5, 2015, at 3:41 AM, Ankit Agrawal wrote:
Hi
I am analysing a protein molecule (MD simulations). I wrote a python script to calculate inertia for the protein and ran it over 40000 frames. I got a data in log file but it is creating a ellipsoid around the molecule.
I want 3 axes of the ellipsoid to be shown in that protein molecule not the ellipsoid as it covers the whole protein molecule. So that I can easily visualize the movement of all 3 axes of ellipsoid over the 40000 frames (means how the axes are changing during trajectory run).
Thanks
Ankit
Department of Chemical Science
IISER-Mohali, India
_______________________________________________
Chimera-users mailing list
Chimera-users@cgl.ucsf.eduhttp://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users