Date: Wed, 12 Feb 2003 13:40:28 -0800 (PST)
From: Thomas Cheatham
Subject: Re: The image command of ptraj of Amber 6 is not working.


> I did try the image command of ptraj in Amber 6. But it is not working.
> And the transform command of rdparm in Amber 6 is obsolete. So is there
> any way else that I can put the outside atoms back? Thanks.

What do you mean that the "image command of ptraj" is not working? Can
you give more specifics as to how it is failing? Are waters not coming
back in? Is your molecule broken into multiple chunks? Is ptraj not
properly recognizing your prmtop as truncated octahedron?

The "image" command of ptraj has been extensively tested/used and although
there could be bugs, likely the problem is a peculiarity of your
particular system or command usage. The image commands are rather
powerful in that you can image separate regions in different manners.

You can either e-mail me or the list with more specifics to help me
understand how I can help you get it to work properly...



Basic usage issues:

-- how many solute molecules do you have? If more than one, center the
first molecule to the origin, image, and then center the whole solute and
image again. This will (in most cases) bring separated solutes back
together. If it does not, imaging will be more tricky. I have only
encounted problems in systems with really minimal solvent (such that a
molecule spans the entire box interacting with its images).

-- if using a non-rectangular shape (like truncated octahedron) decide if
you want the triclinic (slated rectangle) or more familiar (spherical)
shape (i.e. add familiar keyword).

-- make sure if you are stripped solvent to add "nobox" to the trajout
command.

So, assume I have a 10-mer DNA duplex, :1-10 on the first strand and
:11-20 on the second then the script I would run would look like:

trajin mdcrd.1
trajin mdcrd.2
trajout mdcrd.imaged

center :1-10 mass origin
image origin center
center :1-20 mass origin
image origin center familiar

rms first mass out rms :1-20


Good luck in tracking the issue down. Also if you are using an older
version of ptraj, perhaps imaging is broken within it. Up to date
versions are available at http://www.chpc.utah.edu/~cheatham/software.html
and hopefully someday soon when I am able via the AMBER WWW page.


Thomas E. Cheatham, III (Assistant Professor) College of Pharmacy