AMBER Archive (2004)

Subject: Re: AMBER: Questions about MD sampling

From: Thomas E. Cheatham, III (
Date: Tue Oct 12 2004 - 11:12:18 CDT

> I do not know why the sander job crashed (Is there a way to check?). I
> got a kind of error message "pipe broken". (I found it frequently
> happened after the job runing up to three days). Seems something wrong

If you are running in a queue environment, you can check the output logs
created. If you are running manually, when you start the job, pipe the
output and stderr to a file so you can see any error messages (if any).

        mpirun -np 8 sander -O ... >& error_log &

> with the parallel. I run the sander on a linux cluster with 16
> dual-nodes (AMD1.6GHz). My MD system includes protein (580 residues) and
> up to 25000 water. The 2 ns of MD would take almost 2 weeks in general
> (without competition). Is it normal? (I am using " mpirun -np 32
> $AMBERHOME/exe/sander .....). I feel it is kind of slow. Maybe something

You need to benchmark your particular application in a short run (100 or
500 steps) and determine what is the optimal number of processors. My
guess is that you would be better off running on 8 or 16 processors (or
less) especially if you do not have a fast interconnect (like myrinet).
You can also investigate running with PMEMD which has more optimal
parallel performance.

> docking and ligand binding study. However, the loop which occupies the
> active site in free form, bind tightly forming two H-bonds. I did 2ns MD
> in water also in vaco, and could not find any snapshot that the loop fly
> out of the active site.Maybe it is a challenge task for Amber. I was

In vacuo, the hydrogen bonds will be really strong and likely will not
come apart in short simulation. In water, you have a better chance,
however the time scale for the opening may be longer than is accessible in
MD simulation. Two nanoseconds is not a long time. What you are
proposing is a challenging task for room temperature MD simulation (not
AMBER per se). Options are to try forcing the opening by either putting
in distance restraints to break the hbond, using the new targeted RMSd
methodology in AMBER8, possibly LES on the loop region, or even possibly
higher temperature simulation (like application of replica-exchange

> Finally a small question, I got the B-factors (by residues) from MD
> simulation, and I want to compare it with the experimental B-factors in
> PDB. Is there a script so I can extract the experimental B-facators
> easily from PDB and plot it by residues?

The program ptraj can calculate atomic positional fluctuations or
B-factors. To do this, you need to RMS fit the trajectory to a common
reference frame and use the atomicfluct command.

trajin mdcrd.1
trajin mdcrd.2
rms first mass out backbone_rms.dat @C,CA,N
atomicfluct out bfactor_byres.dat :1-200 byres bfactor

The above assumes the protein is residues 1-200.

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to