AMBER Archive (2006)

Subject: Re: AMBER: simulating a small chain protein in water box

From: Carlos Simmerling (
Date: Wed Nov 08 2006 - 11:50:59 CST

if you don't care about the initial structure then you
can afford to be less careful with equilibration.
I usually solvate in leap then minimize the whole thing, then
do MD without restraints. I find that heating from 0K can cause
trouble so I usually start it at around 200 and heat from there.

if you haven't done any explicit water simulations, pick one from
the tutorials and follow it so you are familiar with pressure
coupling, etc. Then do your own.

Be aware that getting the "right" structure in explicit water is
very challenging when you start from something far away.
Doing it in normal MD is almost impossible.
You might look at a recent article that I published in JMB
(as well as articles by others, of course)

**Wickstrom, L., Okur, A., Song, K., Hornak, V., Raleigh, D. and
Simmerling, C., /“The Unfolded State of the Villin Headpiece Helical
Subdomain: Computational Studies of the Role of Locally Stabilized
Structure”/, J. Mol. Biol., 360:1094-1107 (2006).

good luck!

Dave, Sonya wrote:

> Hi,
> I want to simulate the structure of a small chain of amino acids, de
> novo. I actually know this chain is part of an alpha helix in a larger
> protein. I ran the simulation, per the tutorial about TRP cage.
> However, the structure does not look anything like I expect. I think
> this is because I am running in vacuum (the original protein also
> looks improper simulated in vacuum). As such, I want to simulate the
> chain in explicit solvent. How would I go about this? Specifically,
> Can I put the water box around the result of my vacuum simulation,
> rather then the straight chain? Would you suggest doing that?
> Do I need to start from 0 K, and slowly heat up the molecule? In the
> vacuum simulation, the molecule was slowly heated to 300K, so do I
> need to do it again for water simulation?
> Do I need to do all the steps as for simulating a large protein in
> explicit solvent? That is, first minimize the water box, then put weak
> restraints on the protein, then put no restraints on the protein?
> How do I determine the input files, if I do need to do all those
> simulations? Should I use ones similar to those I used for large
> protein explicit simulations? Or should I use the reheating input files?
> This would be the first time i'm doing a simulation without basically
> following some tutorial (with minor modifications).
> Thank you,
> Sonya

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