AMBER Archive (2005)

Subject: Re: AMBER: restart coordinate file

From: Bill Ross (ross_at_cgl.ucsf.edu)
Date: Thu Nov 10 2005 - 09:42:53 CST


> I am trying to pick up a certain structure from a MD trajectory file to
> work as a restart file for the next round refinement calculation.
> Let's say I have got 200 frames in the trajectory and I want to get
> the restart file using the number 134 frame. The following are my
> input of carnal input file.

By the way, carnal is going to be dropped from amber, so you
may want to start getting familiar with ptraj.

> ...END
>
> But the ouput file 'temp.rst.134' does not have the velocity and the box
> information. There is only xyz coordinates in it. And it is cannot be used for
> the next round of calculation. Could anybody give me some clue how I can get a
> correct restart file for a certain frame from the trajectory file?

You cannot get velocities from mdcrd because it does not contain
them. If you saved mdvel at the same interval, you could extract
the corresponding set of velocities using the same method (since
the file format is the same), and then manually create a restrt
by combining coords & velocities for frame 134 using an editor.

The box info may not be in the mdcrd either - look at the file and
see.. if it is, then there is a carnal bug in not transferring it
to the new restrt. Since carnal won't be fixed (unless someone
likes it enough to take over maintenance), you would have to do
this manually or switch to ptraj. If there is no box info (this
might happen w/ constant volume, not sure), then you would have
to get the box from somewhere else and add it manually.

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