|
|||||||||||||||||||||||||||||||||
AMBER Archive (2001)Subject: ptraj: principal axes
From: Thomas C. Bishop (bishop_at_tulane.edu)
Dear amber users,
I'm having trouble convincing myself that the ptraj action
principal mask
actually aligns the principal axes of the selection with the coordinate axes (this is qhat is suppose to do, right?)
In particular if one picks out a phenol ring (or any ring for that matter) and aligns the carbon atoms one would expect that the ring will end up in say the x-y plane or y-z plane or z-x plane. Granted for such rings two of the principal axes will be close to degenerate so there may be some trouble, but ptraj should be able to handle this.
NOTE: in my copy of action.c the "DoRotation" value is came set to 0 during the principal action.
In a related issue. I am not convinced that the ptraj action
vector name mask {principal [| x| y| z|] } | dipole| mask2|box} [ out filename}
is calculating the principal axes correctly either.
Again a simple test on a ring does not produce 2-vectors in the plane of the ring and one perpendicular to this plane.
Can someone help me figure this out.
Thanks,
Here's basically all my ptraj script contains (other than a trajin )
vector princ :242_at_C1,C2,C3,C4,C5,C10 principal x out analy/lig-px.dat
I would also expect that
trajin A.in.crd
would align the principal axes of the backbone of my protein w/ the x,y,z axes and make a new trajout.
-----------------------------------------
| |||||||||||||||||||||||||||||||||
|