AMBER Archive (2005)

Subject: Re: AMBER: Problem in ptraj

From: Thomas E. Cheatham, III (cheatham_at_chpc.utah.edu)
Date: Wed Feb 02 2005 - 12:34:01 CST

> > All of this sounds very strange to me. Does your average structure calculated on
> > the compound trajectory you made with nobox look "normal"?
>
> I checked this structure and found that it is not correct. So I think that
> there is something wrong witht he compound mdcrd file created but what it
> is I dont know.
...
> I checked the individual mdcrd files and generated the average structure
> using each one separately. I found that that the structures thus generated
> are perfectly all right. Also I generated pdb structures for all the
> frames from individual mdcrd files. and these structures are also all
> correct.
>
> I also tried to generate the pdb structures and average structure from the
> individual mdcrd files using the nobox keyword and these structures were
> exactly the same as that generated without the use of nobox.

If I understand your problem correctly, you see garbage results when you
go back and re-do the analysis using the trajectory file created by ptraj
that is the aggregate of the separate trajectories. This is a bit
confusing.

When you created the "aggregate" trajectory with the commands:

  ptraj mod9.prmtop << EOF
  trajin gbmd-pr.mdcrd
  trajin gbmd-pr-1.mdcrd
  trajin gbmd-pr-2.mdcrd
  trajin gbmd-pr-3.mdcrd
  trajin gbmd-pr-4.mdcrd
  trajin gbmd-pr5.mdcrd
  trajout grp3.mdcrd

The output trajectory (grp3.mdcrd) will, I believe, add box information
(stupidly) even though box information is not present (which I assume is
the case since you named the input trajectories as gbmd*).

If you change the trajout to

  trajout grp3.mdcrd nobox

This, as Kristina mentioned, should create a valid trajectory (assuming
you aren't doing anything else inside like shifting the coordinates, etc.)
that will be properly processed using the same prmtop (mod9.prmtop).

Otherwise I am a bit confused as well...

[Note, this is why this trajectory format is particularly poor since there
is no way to easily detect corrupt frames, correspondence to a given
prmtop, etc.]

Let me know if I can be any more help.

--tom

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu