AMBER Archive (2005)

Subject: Re: AMBER: trajectory alignment using ptraj

From: Wen Li (liw_at_wadsworth.org)
Date: Fri Apr 01 2005 - 15:23:14 CST

Thanks for your response, Dr. Cheatham.

First of all, the aligned frames had completely no overlap with my
reference coordinates. So it is obvious that the frames were not led to
the reference although, as I wrote before, these frames are aligned
together, just to somewhere else.

Secondly, the problem in my case is that I need absolutely to fix the
position of the reference since it is solved by being oriented to other
structures.

It would be a great help if the code can be adjusted to this type of need.

Thanks,
Wen

On Fri, 1 Apr 2005, Thomas E. Cheatham, III wrote:

>
> > In addition to my previous message, I randomly chose a frame from a
> > trajectory, and add this frame to the beginning of the trajectory which I
> > want to align to the added first. The result is that the trajectory is not
>
> Are you sure that the frames are not aligned? (i.e. even though they are
> not at the same location in space, they are relatively aligned?) If not,
> this is a problem; if they are simply translated, this is the expected
> behavior (as implemented). I think the earlier ptraj versions (and maybe
> even the current) translate the center of mass of the reference
> coordinates to the origin (to simplify things). If this centering is
> implicitly occuring (which my quick read of the code suggests), you can
> overcome this by finding the center of mass (of the atom selection) for
> your reference coordinates and manually translate (using the translate
> command) after fitting... Alternatively, you can center @P of the
> reference coordinates to the origin (in a separate ptraj run).
>
> Otherwise, the rms() routine in rms.c would have to be modified to shift
> the transformed coordinates back to the center of mass of the reference
> coordinates. I have to think a bit more about (and test) the implications
> of such a change, but I think it may actually be the right thing to do
> here (i.e the rms code should not implicitly center to the origin but to
> the center of mass of the reference coordinates) and should be a
> relatively simple code change...
>
> --tom
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu