On Mar 9, 2010, at 2:17 PM, Francesco Pietra wrote:
Hello Eric:
I have seen your later correction mail.
I had a bad day, that is little time for chimera and with little
success. Please try to guess were I was faulty.
francesco@tya64:~$ chimera --version
chimera alpha version 1.5 (build 30193) 2010-03-06 09:27:34 GMT
francesco@tya64:~$
This is amd64, at the moment I have no 32bit computer.
<select ligand> selects too much.
<select :residue.number>
<select :residue.name>
do select exactly what I want (which is a small ligand of HEME ligand
of the protein.
The problem was with <define> command
MD-movie...Per-Frame "define centroid :residue.number" ... Apply
then activating the movie gave the same view as not going through the
Per_Frame, i.e., the small ligand does not leave any sphere on the
path. I did not investigate if the sphere are too small to be seen at
the magnification I used, ore have the same color as the background
(black). Very likely, I did some other major mistake.
Not sure what to say. Works for me. I've appended an image of centroids tracing the motion of a small ligand in an RNA binding site. I used a step size of 10 and this 'define' command:
define centroid raiseTool false radius 1.0 color goldenrod sel
"raiseTool false" prevents 'define' from trying to bring up the Axes/Planes/Centroids tool (and update that tool's table) for every centroid created. The other options control the size and color of the centroid. I had the small ligand selected so 'sel' is my atom spec. I tried it a second time with ":35" (which also selects the small ligand) and that worked too.
FYI, "~define" will get rid of all the centroids.
I hope this helps somehow. Are you sure you had the "Interpret script as" part of the per-frame dialog set to "Chimera commands"? I can't imagine that you didn't, since you would have gotten some kind of error if it had been interpreted as Python.
--Eric
