AMBER Archive (2006)

Subject: RE: AMBER: Unstable RMS trajectory

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Tue Oct 17 2006 - 11:40:43 CDT


Dear Sonya,
 
How are you generating an RMSD plot from a minimization? The coordinates of
each minimization frame are not output and thus all you can calculate is the
RMS difference between the minimized structure and the original. I suspect
that you are using the various scripts I provide in the tutorial (like
process_mdout.pl) in an innappropiate fashion. The process_mdout script is
only designed for processing output from MD runs while the rmsd calculation
scripts I provide are designed for processing a plotting rmsd's from a
trajectory file. You woun't have this if you are doing minimization.
 
Note, you need to be very careful if you are using xmgr or xmgrace. If you
use one of the scripts on a file that does not contain the necessary data
you will still get an output file but it will contain only a single column
with the time values. E.g.:
 
0.0
0.1
0.2
0.3
0.4
 
etc.
 
If you then just blindly hand this file off to xmgr or xmgrace it will
happily plot you a graph with a straight line of gradient 1.0 and origin
0,0. This is the danger of using things like a black box. You should double
and even tripple check everything you do. Take a look at all the rmsd data
files you calculated etc and make sure they make sense.
 
You should also rely on your eyes a lot more. Overlay the initial and final
structures and look for any obvious differences. If there are problems you
are likely to spot it here.
 
On a second front make sure you are running your imlicit solvent simulation
correctly. That is you are not running just a box of solvent in vacuum. This
is easy to do if you don't set 'ntb' correctly. If you run a solvated box
without periodic boundaries your water molecules will quickly boil off into
space which if you calculate an RMSD vs time for the whole system will look
much like a steadily increasing line. You essentially see the effect of
entropy on your gas phase system. I don't know if you did this by mistake as
you didn't post your input files but it is something worth checking.
 
I hope this helps.
 
All the best
Ross

/\
\/
|\oss Walker

| HPC Consultant and Staff Scientist |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk <http://www.rosswalker.co.uk/> | PGP Key
available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

 

  _____

From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf Of
Dave, Sonya
Sent: Tuesday, October 17, 2006 09:04
To: amber_at_scripps.edu
Subject: AMBER: Unstable RMS trajectory

Hi,
 
I am minimizing a protein in water, using explicit solvent. I'm following
the input files of the tutorial 1, part 3/4. Except, i am applying it to a
protein.
 
The vacuum minimization works fine, in that RMS stabilizes and EP tot
decreases and stabilizes. However, for the water explicit minimizations,
the RMS shoots off to high values, in a smooth continually increasing
fashion. I am using the result of vacuum simulation PDB as the input for
all my water simulations (i am first fixing the protein, the removing
restraints, per the tutorial).
 
I am plotting potential energy and RMS during the simulation. The PE
initially spiked up, but now it's is decreasing and stabilizing (at a
highly negative value). The RMS, however, continues to rise.
 
Does this always mean there is something wrong with my simulation? Or do I
have to make movies of the files to be sure? If it means something is
wrong, how do i decide what to change to make it right?
 
Thank you,
Sonya Dave'

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