Re: AMBER: Torsion Restraint on Glycosidic Bond

From: Thomas Cheatham III <>
Date: Fri, 16 Feb 2007 14:32:30 -0700 (Mountain Standard Time)

> I am refining one of my models trying to hold a NA base in the syn conformation.
> I created an RST file and followed the protocol listed in the DNA tutorial. I
> applied my restraint to a NA structure that I had already minimized and then ran
> the annealing step. No matter what values I assign to the restraint, it doesn't
> seem to affect the particular base that much. It still ends up in the normal
> anti conformation. Am I going about this the right way?

Without further details, it is hard to answer your question directly.

(1) restraints may not be properly applied (however since you see "some"
difference in the structure, likely a restraint is turned on). If you are
optimizing by MD and it is not moving, likely the restraint is not
properly applied.

(2) the anti is a local minima even with the restraint on, i.e. if you are
simply optimizing by minimization, the structure may get trapped in the
local minima of the anti state despite the fact that the restraint energy
is high (as the gradient may be zero). One way to *possibly* overcome
this is in minimization with a large initial step (such that it jumps over
the barrier) such as with conjugate gradient and a large initial step
size. Better, you could do dynamics at a finite temperature, say 300K for
a short time in vacuo and then minimize, or you could rotate the molecule
by hand to flip chi with a graphic programs (like chimera) and then create
a restart for this to put the initial angle closer to syn.

good luck!

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to
Received on Sun Feb 18 2007 - 06:07:37 PST
Custom Search