AMBER Archive (2009)

Subject: Re: [AMBER] Can bonded pair of atoms be restrainted by NMR distance restraint?

From: Carlos Simmerling (carlos.simmerling_at_gmail.com)
Date: Sun Feb 01 2009 - 11:55:54 CST


yes a distance restraint can work like this, but you have lots of other
things defined in your inputs so it might help for you to remove some or all
and troubleshoot to see if a combination is the problem, for example, add
your restraint with standard MD (no belly, don't change REST, don't use
inlty, etc). also you probably don't want to use a restraint on something
that has the harmonic bond, or these two terms will "fight" each other.
perhaps you can remove the bond in leap and then use your restraint.

you probably want r1 to not match r2, but if you clearly understand what the
flatwell potential looks like, and what the energies/forces are in the
various regions between the r values, then maybe this is ok( but it's hard
to help since you don't say what distances you want it to adopt and what the
bond term wants it to be). if the initial structure is less than 3A, then
probably your problem is that r1 = r2 so there is no force.

check carefully to see what sander says in the mdout file after reading the
restraint- does is list what you think it should?

On Sun, Feb 1, 2009 at 2:51 AM, liu junjun <ljjlp03_at_gmail.com> wrote:

> Hello everybody,
>
> I am trying to put distance restraint on a pair of bonded atoms. I
> defined the two atoms, which are actually bonded to each other in the
> intial strcture, in the NMR distance restraint file but it looks like
> the restraint force is not applied on the distance I defined. I set a
> long distance (3 angstrom) as the target and a huge force constant in
> distance restraint input file, which should end up with huge penalty
> on the distance I defined. But the restraint output file does not
> generate any penalty and the minimized structure does not show any
> changes on the distance I defined. The format for input files should
> be OK because I successfully restrained other distances of non-bonded
> pair of atoms with similar input files. I am wondering if NMR distance
> restraint can not be used on bonded atoms or not? If the answer is no,
> then can someone give me a suggestion on restraining a specifically
> pair of bonded atoms? BTW, I am using AMBER9.
>
> here is my input files:
> === sander input ====
> # Control section
> &cntrl
> ntpr = 5,
> cut = 10.0,
> ntb = 0,
> maxcyc = 10,
> ncyc = 20,
> drms=0.05,
> ntr = 0,
> imin = 1, ibelly=1,
> bellymask="@C32 < @20.0 & !(@C32|:197_at_OG)"
> nmropt=1,ipnlty=1
> &end
> &wt type='REST',istep1=1,istep2=100000,
> value1=1.0, value2=1.0,&end
> &wt type='END' /
> LISTOUT=RST.out
> DISANG=dis.RST
> =================
>
> ======= NMR distance restraint file ====
> &rst IRESID=0, iat=8385,8384, r3=3.04,
> r4=3.1, r1=3.04, r2=3.04, rk2=8000, rk3=8000,/
> &end
> ===============================
>
>
> ======= output file for NMR distance restraint ==========
>
> ------------------------------------------------------------------------------
>
>
> Final Restraint Analysis for coords: min1.rst
>
>
> Restraints, deviations, and energy contributions: pencut = 0.10
>
>
> ------------------------------------------------------------------------------
> First atom Last atom curr. value target deviation penalty
>
> ------------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> ====================================
>
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber