Print

Print
I just realized a simplification wherein you can skip quaternions. My guess is you just want a rigid body to sit there and rotate without defined starting and ending orientations. Here is what you do: you decide how fast you want it to rotate, in terms of frames, then you figure out the rotation per frame. You can think of this as an incremental axis and angle rotation. Then you describe that increment as a 3x3 rotation matrix. This conversion is not that difficult. Scipy I think can do this without any thought from the programmer. This 3x3 rotation transformation can get plugged straight into the pymol "transformation matrix". Then, between frames, you just keep applying that transformation and it will rotate dutifully.

If the axis doesn't intersect the origin (e.g. your rigid body is not centered at the origin, etc) you apply the transformation to center it, apply the rotation, then move the rigid body back, then take the frame. Repeat.

Probably such a thing would be trivial to add to emovie since its open sourced. The parameters would be (x0,y0,z0) and (x1,y1,z1) describing the arbitrary rotation axis, and d, describing the number of degrees to rotate per frame. The bigger the d, the faster.

James



On Aug 1, 2008, at 2:06 PM, Jerry McCully wrote:

Hi , all:

     Sorry for this off-topic question.

     When you make movies, Pymol usually allows you rotate molecules along x, y, or z axis.

     The view point is always from Z axis.

    Does anyone have this experience of rotating molecules along any other specified axis and keeping the same view point as usual?

    Thanks a lot.

Jerry

Use video conversation to talk face-to-face with Windows Live Messenger. Get started.

--
James Stroud
UCLA-DOE Institute for Genomics and Proteomics
611 Charles E. Young Dr. S.
Los Angeles, CA  90095